PhD, 2000

Department of Hydrology and Water Resources, University of Arizona, Tucson, Arizona, USA
Major: Engineering Hydrology
Minor: Applied Mathematics
Dissertation title: Numerical inverse interpretation of pneumatic tests in unsaturated fractured tuffs at the Apache Leap Research Site
Advisor: Regents Professor Dr. Shlomo P. Neuman

MEng, 1989

Department of Hydrogeology and Engineering Geology, Institute of Mining and Geology, Sofia, Bulgaria
Major: Hydrogeology
Minor: Engineering Geology
Dissertation title: Hydrogeological investigation in applying the Vyredox method for groundwater decontamination
Advisor: Professor Dr. Pavel P. Pentchev

• machine learning; unsupervised/supervised/deep; nonnegative matrix/tensor factorization (decomposition); feature extraction; source identification; blind source separation; big-data analytics;
• evaluation of information content; data-worth analyses; big-data analytics; extraction of actionable knowledge/information
• model analyses; parameter estimation; sensitivity analyses; uncertainty quantification; risk assessment; decision support;
• model abstraction; model reduction; reduced order modeling; support vector machines; support vector regression; model selection; model ranking;
• optimal experimental design; optimal data acquisition;
• General Information Theory (GIT); Fuzzy sets; Bayesian techniques; Information Gap Decision Theory; Bayesian Information Gap (BIG) Decision Theory;
• optimization (single/multi-objective, global/local techniques, Levenberg-Marquardt and Particle Swarm methods);
• characterization of flow medium heterogeneity; high-resolution stochastic imaging (tomography); stochastic inverse analysis;
• scale effects in properties of porous/fractured media; Lévy (alpha-stable) distributions; fractal formalism;
• analytical and numerical techniques for simulation of flow/transport in saturated/unsaturated, porous/fractured media;
• simulation of flow/transport in saturated/unsaturated, porous/fractured media; analytical and numerical modeling techniques
• regional/site scale conceptual and numerical modeling; evaluation of conceptual model uncertainties;
• subsurface fluid dynamics and contaminant transport; well hydraulics; exploration and protection of groundwater resources; design of groundwater-supply systems; capture-zone analyses;
• high-performance computing; quantum computing; code development

• Lopez, C.A, Vesselinov, V.V., Gnanakaran, S., Alexandrov, B.S., Unsupervised Machine Learning for Analysis of Coexisting Lipid Phases and Domain Growth in Biological Membranes, PLoS ONE, 2019, (submitted).
• O'Malley, D., Golden, J.K., Vesselinov, V.V., Learning to regularize with a variational autoencoder for hydrologic inverse analysis, arXiv:1906.02401v1, 2019, (submitted). PDF
• Vesselinov, V.V., Mudunuru, M., Karra, S., O'Malley, D., Alexandrov, B.S., Unsupervised Machine Learning Based on Non-Negative Tensor Factorization for Analyzing Reactive-Mixing, 10.1016/j.jcp.2019.05.039, Journal of Computational Physics, 2019. PDF
• Vesselinov, V.V., Alexandrov, B.S., O'Malley, D., Nonnegative Tensor Factorization for Contaminant Source Identification, Journal of Contaminant Hydrology, 10.1016/j.jconhyd.2018.11.010, 2018. PDF
• O'Malley, D., Vesselinov, V.V., Alexandrov, B.S., Alexandrov, L.B., Nonnegative/binary matrix factorization with a D-Wave quantum annealer, PlosOne, 10.1371/journal.pone.0206653, 2018. PDF
• Telfeyan, K., Migdisov, A.A., Pandey, S., Vesselinov, V.V., Reimus, P.W., Long-term stability of dithionite in alkaline anaerobic aqueous solution, Applied Geochemistry, 10.1016/j.apgeochem.2018.12.015, 2018. PDF
• Stanev, V., Vesselinov, V.V., Kusne, A.G., Antoszewski, G., Takeuchi,I., Alexandrov, B.A., Unsupervised Phase Mapping of X-ray Diffraction Data by Nonnegative Matrix Factorization Integrated with Custom Clustering, Nature Computational Materials, 10.1038/s41524-018-0099-2, 2018. PDF
• Iliev, F.L., Stanev, V.G., Vesselinov, V.V., Alexandrov, B.S., Nonnegative Matrix Factorization for identification of unknown number of sources emitting delayed signals PLoS ONE, 10.1371/journal.pone.0193974. 2018. PDF
• Stanev, V.G., Iliev, F.L., Hansen, S.K., Vesselinov, V.V., Alexandrov, B.S., Identification of the release sources in advection-diffusion system by machine learning combined with Green function inverse method, Applied Mathematical Modelling, 10.1016/j.apm.2018.03.006, 2018. PDF
• Qian, E., Peherstorfer, B., O'Malley, D., Vesselinov, V.V., Wilcox, K., Multifidelity Monte Carlo Estimation of Variance and Sensitivity Indices, SIAM Journal on Uncertainty Quantification, 10.1137/17M1151006, 2018. PDF
• Lu, Z., Vesselinov, V.V., Lei, H., Identifying Arbitrary Parameter Zonation using Multiple Level Set Functions, Journal of Computational Physics, 10.1016/j.jcp.2018.03.016, 2018. PDF
• Hansen, S.K., He, J., Vesselinov, V.V., Characterizing the impact of model error in hydrologic time series recovery inverse problems, 10.1017/j.advwatres.2017.146.R2, Advances in Water Resources, 2018. PDF
• Lin, Y., O'Malley, D., Vesselinov, V.V., Gutrie, G.D, Coblentz, D., Randomization in Characterizing the Subsurface, SIAM News, 2018. PDF
• Hansen, S.K., Haslauer, C.P., Cirpka, O.A., Vesselinov, V.V., Direct Breakthrough Curve Prediction from Statistics of Heterogeneous Conductivity Fields, Water Resources Research, 10.1002/2017WR020450, 2018. PDF
• Vesselinov, V.V., Mudunuru. M., Karra, S., O'Malley, D., Alexandrov, Unsupervised Machine Learning Based on Non-negative Tensor Factorization for Analysis of Filed Data and Simulation Outputs, Computational Methods in Water Resources (CMWR), Saint-Malo, France, 10.13140/RG.2.2.27777.92005, 2018. PDF
• Vesselinov, V.V., O'Malley, D., Alexandrov, B.S., Contaminant source identification using semi-supervised machine learning, Journal of Contaminant Hydrology, 10.1016/j.jconhyd.2017.11.002, 2017. PDF
• Hansen, S.K., Pandey, S., Karra, S., Vesselinov, V.V., CHROTRAN 1.0: A mathematical and computational model for in situ heavy metal remediation in heterogeneous aquifers, Geoscientific. Model Development, 10.5194/gmd-10-4525-2017, 10, 4525–4538, 2017. PDF
• Lin, Y, Le, E.B, O'Malley, D., Vesselinov, V.V., Bui-Thanh, T., Large-Scale Inverse Model Analyses Employing Fast Randomized Data Reduction, Water Resources Research, 10.1002/2016WR020299RRR, 2017. PDF
• Hansen, S.K., Vesselinov, V.V., Local equilibrium and retardation revisited, Groundwater, 10.1111/gwat.12551, 2017. PDF
• Hansen, S.K., Vesselinov, V.V., Reimus, P., Lu, Z., Inferring subsurface heterogeneity from push-drift tracer tests, Water Resources Research, 10.1002/2017WR020852R, 2017. PDF
• Bakarji, J., Vesselinov, V.V., O’Malley, D., Agent-based Socio-hydrological Hybrid Modeling for Water Resource Management, Water Resources Management, 10.1007/s11269-017-1713-7, 2017. PDF
• Zhang, X., Sun, A.Y., Duncan, I.J., Vesselinov, V.V., Two-Stage Fracturing Wastewater Management in Shale Gas Development, Ind. Eng. Chem. Res., 10.1021/acs.iecr.6b03971, 2017. PDF
• Zhang, X., Vesselinov, V.V., Integrated Modeling Approach for Optimal Management of Water, Energy and Food Security Nexus, Advances in Water Resources, 10.1016/j.advwatres.2016.12.017, 2017. PDF
• O'Malley, D., Vesselinov, V.V., ToQ.jl: A high-level programming language for D-Wave machines based on Julia. IEEE High Performance Extreme Computing, 10.1109/HPEC.2016.7761616, 2016. PDF
• Lin, Y, O'Malley, D., Vesselinov, V.V., A computationally efficient parallel Levenberg-Marquardt algorithm for highly parameterized inverse model analyses, Water Resources Research, 10.1002/2016WR019028, 2016. PDF
• Hansen, S.K., Berkowitz, B., Vesselinov, V.V., O'Malley, D., Karra, S., Push-pull tracer tests: their information content and use for characterizing non-Fickian, mobile-immobile behavior, Water Resources Research, 10.1002/2016WR018769RR, 2016. PDF
• Zhang, X., Vesselinov, V.V., Energy-Water Nexus: Balancing the Tradeoffs between Two-Level Decision Makers Applied Energy, Applied Energy, 10.1016/j.apenergy.2016.08.156, 2016. PDF
• Hansen, S.K., Vesselinov, V.V., Contaminant point source localization error estimates as functions of data quantity and model quality, 10.1016/j.jconhyd.2016.09.003, 2016. PDF
• Throckmorton, H., Newman, B., Heikoop, J., Perkins, G., Feng, X., Graham, D., O'Malley, D., Vesselinov, V.V., Young, J., Wullschleger, S., Wilson, C., Active layer hydrology in an arctic tundra ecosystem: quantifying water sources and cycling using water stable isotopes, Hydrological Processes, 10.1002/hyp.10883, 2016. PDF
• Grasinger, M., O'Malley, D., Vesselinov, V.V., Karra, S., Decision Analysis for Robust CO2 Injection: Application of Bayesian-Information-Gap Decision Theory, International Journal of Greenhouse Gas Control, 10.1016/j.ijggc.2016.02.017, 2016. PDF
• Mattis, S.A., Butler, T.D. Dawson, C.N., Estep, D., Vesselinov, V.V., Parameter estimation and prediction for groundwater contamination based on measure theory, Water Resources Research, 10.1002/2015WR017295, 2015. PDF
• O’Malley, D., Vesselinov, V.V., Bayesian-Information-Gap Decision Theory (BIG-DT) with an application to CO2 sequestration, Water Resources Research, 10.1002/2015WR017413, 2015. PDF
• Lu, Z., Vesselinov, V.V., Analytical Sensitivity Analysis of Transient Groundwater Flow in a Bounded Model Domain using Adjoint Method, Water Resources Research, 10.1002/2014WR016819, 2015. PDF
• Barajas-Solano, D. A., Wohlberg, B., Vesselinov, V.V., Tartakovsky, D. M., Linear Functional Minimization for Inverse Modeling, Water Resources Research, 10.1002/2014WR016179, 2015. PDF
• O’Malley, D., Vesselinov, V.V., Cushman, J.H., Diffusive mixing and Tsallis entropy, Physical Review E, 10.1103/PhysRevE.91.042143, 2015. PDF
• O’Malley, D., Vesselinov, V.V., A combined probabilistic/non-probabilistic decision analysis for contaminant remediation, Journal on Uncertainty Quantification, SIAM/ASA, 10.1137/140965132, 2014. PDF
• Vesselinov, V.V., O'Malley, D., Katzman, D., Robust Decision Analysis for Environmental Management of Groundwater Contamination Sites, In Vulnerability, Uncertainty, and Risk Quantification, Mitigation, and Management (ed. Michael Beer, Siu-Kui Au, and Jim W. Hall), 2916 pp, ISBN: 9780784413609, 10.1061/9780784413609.197, 2014. Link
• O’Malley, D., Vesselinov, V.V., Cushman, J.H., A Method for Identifying Diffusive Trajectories with Stochastic Model, Journal of Statistical Physics, Springer, 10.1007/s10955-014-1035-6, 2014. PDF
• Alexandrov, B., Vesselinov, V.V., Blind source separation for groundwater level analysis based on non-negative matrix factorization, Water Resources Research, 10.1002/2013WR015037, 2014. PDF
• O’Malley, D., Vesselinov, V.V., Analytical solutions for anomalous dispersion transport, Advances in Water Resources, 10.1016/j.advwatres.2014.02.006, 2014. PDF
• Heikoop, J.M., Johnson, T.M., Birdsell, K.H., Longmire, P., Hickmott, D.D., Jacobs, E.P., Broxton, D.E., Katzman, D., Vesselinov, V.V., Ding, M., Vaniman, D.T., Reneau, S.L., Goering, T.J., Glessner, J., Basu, A., Isotopic evidence for reduction of anthropogenic hexavalent chromium in Los Alamos National Laboratory groundwater, Chemical Geology, 10.1016/j.chemgeo.2014.02.022, 2014.
• Freedman, V.L., Chen, X., Finsterle, S., Freshley, M., Gorton, I., Gosink, L., Keating, E., Lansing, C., Moeglein W., Murray C., Pau, G., Porter, E., Purohit, S., Rockhold, M., Schuchardt, K., Sivaramakrishnan, C., Vesselinov, V.V., Waichler, S., A high-performance workflow system for subsurface simulation, Environmental Modelling & Software, 55, pp. 176-189, 10.1016/j.envsoft.2014.01.030, 2014. PDF
• O’Malley, D., Vesselinov, V.V., Groundwater remediation using the information gap decision theory, Water Resources Research, 10.1002/2013WR014718, 2014. PDF
• Harp, D.R., Vesselinov, V.V., Accounting for the influence of aquifer heterogeneity on spatial propagation of pumping drawdown, Journal of Water Resource and Hydraulic Engineering, 2(3), pp. 65-83, 2013. PDF
• Vesselinov, V.V., Katzman, D., Broxton, D., Birdsell, K., Reneau, S., Vaniman, D., Longmire, P., Fabryka-Martin, J., Heikoop, J., Ding, M., Hickmott, D., Jacobs, E., Goering, T., Harp, D.R., Mishra, P., Data and Model-Driven Decision Support for Environmental Management of a Chromium Plume at Los Alamos National Laboratory (LANL), Waste Management Symposium 2013, Session 109: ER Challenges: Alternative Approaches for Achieving End State, Phoenix, AZ, http://wmsym.org, 2013. PDF
• Vesselinov, V.V., Pau, G., Finsterle, S, AGNI: Coupling Model Analysis Tools and High-Performance Subsurface Flow and Transport Simulators for Risk and Performance Assessments, Computational Methods in Water Resources (CMWR 2012), 2012. PDF
• Vesselinov, V.V., Harp, D., Adaptive hybrid optimization strategy for calibration and parameter estimation of physical models, Computers & Geosciences, 10.1016/j.cageo.2012.05.027, 2012. PDF
• Harp, D., Vesselinov, V.V., Contaminant remediation decision analysis using information gap theory, Stochastic Environmental Research and Risk Assessment (SERRA), 10.1007/s00477-012-0573-1, 2012. PDF
• Mishra, P.K., Gupta, H.V., Vesselinov, V.V.; On simulation and analysis of variable-rate pumping tests, Ground Water, 10.1111/j.1745-6584.2012.00961.x, 2012. PDF
• Mishra, P.K., Vesselinov, V.V., Kuhlmna, K.L.; Saturated–unsaturated flow in a compressible leaky-unconfined aquifer, Advances in Water Resources, 10.1016/j.advwatres.2012.03.007, 2012. PDF
• Mishra, P.K., Vesselinov, V.V., Neuman, S.P.; Radial flow to a partially penetrating well with storage in an anisotropic confined aquifer, Journal of Hydrology, 10.1016/j.jhydrol.2012.05.010, 2012. PDF
• Harp, D., Vesselinov, V.V., An agent-based approach to global uncertainty and sensitivity analysis, Computers & Geosciences, 10.1016/j.cageo.2011.06.025, 2012. PDF
• Harp, D., Vesselinov, V.V., Analysis of hydrogeological structure uncertainty by estimation of hydrogeological acceptance probability of geostatistical models, Special issue of Uncertainty Quantification (invited), Advances in Water Resources, 10.1016/j.advwatres.2011.06.007, 2012. PDF
• Mishra, P.K., Vesselinov, V.V., Unified Analytical Solution for Radial Flow to a Well in a Confined Aquifer, arXiv:1110.5940, 2011. PDF
• Vesselinov, V.V., Harp, D., Decision support based on uncertainty quantification of model predictions of contaminant transport, CMWR 2010: XVIII International Conference on Water Resources, J. Carrera (Ed), CIMNE, Barcelona 2010. PDF
• Harp, D., Vesselinov, V.V., Identification of Pumping Influences in Long-Term Water Level Fluctuations, Ground Water, 10.1111/j.1745-6584.2010.00725.x., 2010. PDF
• Morales-Casique, E, Neuman, S.P., Vesselinov, V.V., Maximum Likelihood Bayesian Averaging of air flow models in unsaturated fractured tuff using Occam and variance windows, Special issue of Stochastic Environmental Research and Risk Assessment (SERRA) Journal celebrating 70th anniversary of Shlomo P Neuman, vol. 24, 10.1007/s00477-010-0383-2, 2010. PDF
• Harp, D., Vesselinov, V.V., Stochastic inverse method for estimation of geostatistical representation of hydrogeologic stratigraphy using borehole logs and pressure, invited, Special issue of Stochastic Environmental Research and Risk Assessment (SERRA) Journal celebrating 70th anniversary of Shlomo P Neuman, vol. 24, 10.1007/s00477-010-0403-2, 2010. PDF
• Vrugt, J., Stauffer, P., Wöhling, Th., Robinson, B., Vesselinov, V.V., Inverse Modeling of Subsurface Flow and Transport Properties Using Recent Advances in Global Optimization, Parallel Computing and Sequential Data Assimilation, Vadose Zone Journal, pp 843-864, 2008. PDF
• Morales-Casique, E., Neuman, S.P., Vesselinov, V.V., Maximum likelihood Bayesian averaging of air flow models in unsaturated fractured tuff, pp.70-75, IAHS Publication 320, ISBN 978-1-901502-49-7, 2008. PDF
• Harp, D., Dai, Z., Wolfsberg, A., Vrugt, J., Robinson, B., Vesselinov, V.V., Aquifer structure identification using stochastic inversion. Geophysical Research Letters L08404, 10.1029/2008GL033585, 2008. PDF
• Vesselinov, V.V., Uncertainties In Transient Capture-Zone Estimates, Computational Methods in Water Resources XVI, (edited by P. Binning, P. Engesgaard, H. Dahle, G. Pinder & W. Gray), Balkema, Rotterdam, ISBN 90-5809-124-4, pp. 307-314, 2006. PDF
• Vrugt, J.A, Robinson, B.A., Vesselinov, V.V., Improved Inverse Modeling in Geophysics: Combined Parameter and State Estimation, Geophysical Research Letters, v.32, L18408, 10.1029/2005GL023940, 2005. PDF
• Vesselinov, V.V., Estimation of parameter uncertainty using inverse model sensitivities, Computational Methods in Water Resources XV (CMWR 2004) (ed. Miller, C., Farthing, M.W., Gray, W.G., Pinder, G.), Elsevier, ISBN 0-444-51839-8, pp. 508-514, 10.1016/S0167-5648(04)80139-4, 2004. PDF

More publications are available at Google Scholar, ResearchGate, and Academia.edu.

• Vesselinov, V.V., Physics-Informed Machine Learning Methods for Data Analytics and Model Diagnostics, M3 NASA DRIVE Workshop, Los Alamos, 2019. PDF
• Vesselinov, V.V., Novel Unsupervised Machine Learning Methods for Data Analytics and Model Diagnostics, Machine Learning in Solid Earth Geoscience, Santa Fe, 2019. PDF
• Vesselinov, V.V., Unsupervised Machine Learning Methods for Feature Extraction, New Mexico Big Data & Analytics Summit, nmbdas.com, Albuquerque, 2019. PDF
• Vesselinov, V.V., Novel Machine Learning Methods for Extraction of Features Characterizing Datasets and Models, AGU Fall meeting, Washington D.C., 2018. PDF
• Vesselinov, V.V., Novel Machine Learning Methods for Extraction of Features Characterizing Complex Datasets and Models, Recent Advances in Machine Learning and Computational Methods for Geoscience, Institute for Mathematics and its Applications, University of Minnesota, 10.13140/RG.2.2.16024.03848, 2018. PDF
• Vesselinov, V.V., Mudunuru. M., Karra, S., O'Malley, D., Alexandrov, Unsupervised Machine Learning based on Nonnegative Matrix/Tensor Factorization, World Congress on Computational Mechanics (WCCM), New York, NY, 2018 (invited)
• Vesselinov, V.V., Mudunuru. M., Karra, S., O'Malley, D., Alexandrov, Unsupervised Machine Learning Based on Non-negative Tensor Factorization for Analysis of Filed Data and Simulation Outputs, Computational Methods in Water Resources (CMWR), Saint-Malo, France, 10.13140/RG.2.2.27777.92005, 2018. PDF
• O'Malley, D., Vesselinov, V.V., Alexandrov, B.S., Alexandrov, L.B., Nonnegative/binary matrix factorization with a D-Wave quantum annealer PDF
• Vesselinov, V.V., O'Malley, D., Alexandrov, B., Novel Robust Machine Learning Methods for Identification and Extraction of Unknown Features in Complex Real-world Data Sets, Society for Industrial and Applied Mathematics (SIAM) Uncertainty Quantification, Garden Grove, CA, 2018, (invited).
• Vesselinov, V.V., O'Malley, D., Alexandrov, B., Unsupervised Machine Learning Based on Tensor Factorization, International Society for Porous Media (INTERPORE), New Orleans, LA, 10.13140/RG.2.2.15195.00807, 2018. PDF
• Vesselinov, V.V., O'Malley, D., Alexandrov, B., Uncertainty quantification and experimental design based on unsupervised machine learning identification of contaminant sources and groundwater types using hydrogeochemical data, AGU Fall Meeting, New Orleans, LA, 2017 PDF
• O'Malley, D., Vesselinov, V.V., Quo vadis: Hydrologic inverse analyses using high-performance computing and a D-Wave quantum annealer, AGU Fall Meeting, New Orleans, LA, 10.13140/RG.2.2.10161.84320, 2017. PDF
• Lin, Y., Vesselinov, V.V., O'Malley, D., Wohlberg, B., Hydraulic Inverse Modeling using Total-Variation Regularization with Relaxed Variable-Splitting, SIAM Conference on Computational Science and Engineering, Atlanta, GA, 2017. PDF
• Vesselinov, V.V., O'Malley, D., Katzman, D., Decision Analyses for Groundwater Remediation, Waste Management Symposium, Phoenix, AZ, 10.13140/RG.2.2.33273.11367, 2017. PDF PDF
• Vesselinov, V.V., O'Malley, D., Model Analyses of Complex Systems Behavior using MADS, AGU Fall Meeting, San Francisco, CA, 10.13140/RG.2.2.33902.25921, 2016. PDF
• Vesselinov, V.V., O'Malley, D., Alexandrov, B., Moore, B., Reduced Order Models for Decision Analysis and Upscaling of Aquifer Heterogeneity, AGU Fall Meeting, San Francisco, CA, 2016, (invited). PDF
• He, J., Hansen, S.K., Vesselinov, V.V., Analysis of Hydrologic Time Series Reconstruction Uncertainty due to Inverse Model Inadequacy, AGU Fall Meeting, San Francisco, CA, 2016. PDF
• Lu, Z., Vesselinov, V.V., Lei, H., Identifying Aquifer Heterogeneities using the Level Set Method, AGU Fall Meeting, San Francisco, CA, 2016. PDF
• Hansen, S.K., Haslauer, C.P., Cirpka, O.A., Vesselinov, V.V., Prediction of Breakthrough Curves for Conservative and Reactive Transport, AGU Fall Meeting, San Francisco, CA, 2016. PDF
• Zhang, X., Vesselinov, V.V., Bi-Level Decision Making for Supporting Energy and Water Nexus, AGU Fall Meeting, San Francisco, CA, 2016. PDF
• Lin, Y., Vesselinov, V.V., O'Malley, D., Wohlberg, B., Hydraulic Inverse Modeling using Total-Variation Regularization with Relaxed Variable-Splitting, AGU Fall Meeting, San Francisco, CA, 2016. PDF
• Vesselinov, V.V., O'Malley, D., Katzman, D., ZEM: Integrated Framework for Real-Time Data and Model Analyses for Robust Environmental Management Decision Making, Waste Management Symposium, Phoenix, AZ, 10.13140/RG.2.2.29917.67041, 2016. PDF
• Vesselinov, V.V., O'Malley, D., Katzman, D., Model-Assisted Decision Analyses Related to a Chromium Plume at Los Alamos National Laboratory, Waste Management Symposium, Phoenix, AZ, 10.13140/RG.2.2.23206.78404, 2015. PDF
• O'Malley, D., Vesselinov, V.V., Bayesian Information-Gap (BIG) Decision Analysis Applied to a Geologic CO2 Sequestration Problem, AGU Fall Meeting, San Francisco, CA, 2014. PDF
• Cushman, J.H., Vesselinov, V.V., O'Malley, D., Random dispersion coefficients and Tsallis entropy, AGU Fall Meeting, San Francisco, CA, 2014. PDF
• Bakarji, J., O'Malley, D., Vesselinov, V.V., A Social Dynamics Dependent Water Supply Well Contamination Model, LANL Postdoc Research Conference, 10.13140/RG.2.2.17963.90407, 2014. PDF
• Vesselinov, V.V., Alexandrov, B.A, Model-free Source Identification, AGU Fall Meeting, San Francisco, CA, 2014. PDF
• Vesselinov, V.V., Katzman, D., Broxton, D., Birdsell, K., Reneau, S., Vaniman, D., Longmire, P., Fabryka-Martin, J., Heikoop, J., Ding, M., Hickmott, D., Jacobs, E., Goering, T., Harp, D., Mishra, P., Data and Model-Driven Decision Support for Environmental Management of a Chromium Plume at Los Alamos National Laboratory (LANL), Session 109: Environmental Restoration Challenges: Alternative Approaches for Achieving End State, Waste Management Symposium, Phoenix, AZ, February 28, 10.13140/RG.2.2.10414.15686, 2013. PDF
• Vesselinov, V.V., Harp, D., Katzman, D., Model-driven decision support for monitoring network design based on analysis of data and model uncertainties: methods and applications, H32F: Uncertainty Quantification and Parameter Estimation: Impacts on Risk and Decision Making, AGU Fall meeting, San Francisco, December 3-7, 10.13140/RG.2.2.17125.04321, LA-UR-13-20189, 2012, (invited). PDF
• Vesselinov, V.V., et al., AGNI: Coupling Model Analysis Tools and High-Performance Subsurface Flow and Transport Simulators for Risk and Performance Assessments, XIX International Conference on Computational Methods in Water Resources (CMWR 2012), University of Illinois at Urbana-Champaign, June 17-22, 2012. PDF
• Leif Zinn-Bjorkman, L., Numerical Optimization using the Levenberg-Marquardt Algorithm, EES-16 Seminar Series, LA-UR-11-12010, 10.13140/RG.2.2.11253.01760, 2011. PDF
• Harp, D., Vesselinov, V.V., Recent developments in MADS algorithms: ABAGUS and Squads, EES-16 Seminar Series, LA-UR-11-11957, 10.13140/RG.2.2.35579.98082, 2011. PDF
• Vesselinov, V.V., et al., Environmental Management Modeling Activities at Los Alamos National Laboratory (LANL), Department of Energy Technical Exchange Meeting, Performance Assessment Community of Practice, Hanford, April 13-14, 10.13140/RG.2.2.12091.87846, 2010. PDF
• Vesselinov, V.V., Harp, D., Koch, R., Birdsell, K., Katzman, K., Tomographic inverse estimation of aquifer properties based on pressure variations caused by transient water-supply pumping, AGU Meeting, San Francisco, CA, December 15-19, 10.13140/RG.2.2.25513.65126, 2008. PDF
• Vesselinov, V.V., Uncertainties in Transient Capture-Zone Estimates, CMWR 2006 XVI International Conference on Computational Methods in Water Resources, Copenhagen, Denmark, June 18-22, 10.13140/RG.2.2.28869.09446, 2006. PDF

Presentations are also available at SlideShare.net, ResearchGate, and Academia.edu.

• Vesselinov, V.V., Novel Machine Learning Methods for Extraction of Features Characterizing Complex Datasets and Models, Recent Advances in Machine Learning and Computational Methods for Geoscience, Institute for Mathematics and its Applications, University of Minnesota, 2018. Video

Videos are also available on YouTube

• Fate and Transport Investigations Update for Chromium Contamination from Sandia Canyon, LA-UR-08-4702, 2008. PDF
• Pajarito Canyon Investigation Report, LA-UR-08-5852, 2008. PDF
• Decision analysis for addressing groundwater contaminants from the radioactive liquid waste treatment facility released into Mortandad canyon, LA-UR-05-6397, 2005. PDF

More reports are available at LANL electronic public reading room

• Alexandrov, B.S., Vesselinov, V.V., Alexandrov, L.B., Stanev, V., Iliev, F.L., Source identification by non-negative matrix factorization combined with semi-supervised clustering, US20180060758A1

Unsupervised Machine Learning methods are powerful data-analytics tools capable of extracting important features hidden (latent) in large datasets without any prior information. The physical interpretation of the extracted features is done a posteriori by subject-matter experts.

In contrast, supervised Machine Learning methods are trained based on large labeled datasets The labeling is performed a priori by subject-matter experts. The process of deep Machine Learning commonly includes both unsupervised and supervised techniques LeCun, Bengio, and Hinton 2015 where unsupervised Machine Learning are applied to facilitate the process of data labeling.

The integration of large datasets, powerful computational capabilities, and affordable data storage has resulted in the widespread use of Machine Learning in science, technology, and industry.

Recently we have developed a novel unsupervised Machine Learning methods. The methods are based on Matrix/Tensor Decomposition (Factorization) coupled with sparsity and nonnegativity constraints. The method reveals the temporal and spatial footprints of the extracted features.

Tensor Decomposition

A novel unsupervised Machine Learning based on Tensor Decomposition coupled with sparsity and nonnegativity constraints has been applied to extract the temporal and spatial footprints of the features in multi-dimensional datasets in the form of multi-way arrays or tensors. The decomposition (factorization) of a given tensor \(X\) is typically performed by minimization of the Frobenius norm:

$$ \frac{1}{2} ||X-G \otimes_1 A_1 \otimes_2 A_2 \dots \otimes_n A_n ||_F^2 $$

where:

• \(n\) is the dimensionality of the tensor \(X\)
• \(G\) is a "mixing" core tensor
• \(A_1,A_2,\dots,A_n\) are "feature” factors (in the form of vectors or matrices)
• \(\otimes\) is a tensor product applied to fold-in factors \(A_1,A_2,\dots,A_n\) in each of the tensor dimensions

The product \(G \otimes_1 A_1 \otimes_2 A_2 \dots \otimes_n A_n\) is an estimate of \(X\) (\(X_{est}\)).

The reconstruction error \(X - X_{est}\) is expected to be random uncorrelated noise.

\(G\) is a \(n\)-dimensional tensor with a size and a rank lower than the size and the rank of \(X\). The size of tensor \(G\) defines the number of extracted features (signals) in each of the tensor dimensions.

The factor matrices \(A_1,A_2,\dots,A_n\) represent the extracted features (signals) in each of the tensor dimensions. The number of matrix columns equals the number of features in the respective tensor dimensions (if there is only 1 column, the particular factor is a vector). The number of matrix rows in each factor (matrix) \(A_i\) equals the size of tensor X in the respective dimensions.

The elements of tensor \(G\) define how the features along each dimension (\(A_1,A_2,\dots,A_n\)) are mixed to represent the original tensor \(X\).

The Tensor Decomposition is commonly performed using Candecomp/Parafac (CP) or Tucker decomposition models.

Some of the decomposition models can theoretically lead to unique solutions under specific, albeit rarely satisfied, noiseless conditions. When these conditions are not satisfied, additional minimization constraints can assist the factorization.

A popular approach is to add sparsity and nonnegative constraints. Sparsity constraints on the elements of G reduce the number of features and their mixing (by having as many zero entries as possible). Nonnegativity enforces parts-based representation of the original data which also allows the Tensor Decomposition results for \(G\) and \(A_1,A_2,\dots,A_n\) to be easily interrelated Cichocki et al, 2009.

Algorithms called NMFk and NTFk for Matrix/Tensor Factorization (Decomposition) coupled with sparsity and nonnegativity constraints custom k-means clustering has been developed in Julia

NMFk and NTFk codes are available as open-source on GitHub.

Research Papers

• Vesselinov, V.V., Mudunuru, M., Karra, S., O'Malley, D., Alexandrov, B.S., Unsupervised Machine Learning Based on Non-Negative Tensor Factorization for Analyzing Reactive-Mixing, 10.1016/j.jcp.2019.05.039, Journal of Computational Physics, 2019. PDF
• Vesselinov, V.V., Alexandrov, B.S., O'Malley, D., Nonnegative Tensor Factorization for Contaminant Source Identification, Journal of Contaminant Hydrology, 10.1016/j.jconhyd.2018.11.010, 2018. PDF
• O'Malley, D., Vesselinov, V.V., Alexandrov, B.S., Alexandrov, L.B., Nonnegative/binary matrix factorization with a D-Wave quantum annealer, PlosOne, 10.1371/journal.pone.0206653, 2018. PDF
• Stanev, V., Vesselinov, V.V., Kusne, A.G., Antoszewski, G., Takeuchi,I., Alexandrov, B.A., Unsupervised Phase Mapping of X-ray Diffraction Data by Nonnegative Matrix Factorization Integrated with Custom Clustering, Nature Computational Materials, 10.1038/s41524-018-0099-2, 2018. PDF
• Iliev, F.L., Stanev, V.G., Vesselinov, V.V., Alexandrov, B.S., Nonnegative Matrix Factorization for identification of unknown number of sources emitting delayed signals PLoS ONE, 10.1371/journal.pone.0193974. 2018. PDF
• Stanev, V.G., Iliev, F.L., Hansen, S.K., Vesselinov, V.V., Alexandrov, B.S., Identification of the release sources in advection-diffusion system by machine learning combined with Green function inverse method, Applied Mathematical Modelling, 10.1016/j.apm.2018.03.006, 2018. PDF
• Vesselinov, V.V., O'Malley, D., Alexandrov, B.S., Contaminant source identification using semi-supervised machine learning, Journal of Contaminant Hydrology, 10.1016/j.jconhyd.2017.11.002, 2017. PDF
• Alexandrov, B., Vesselinov, V.V., Blind source separation for groundwater level analysis based on non-negative matrix factorization, Water Resources Research, 10.1002/2013WR015037, 2014. PDF

Presentations

• Vesselinov, V.V., Novel Machine Learning Methods for Extraction of Features Characterizing Datasets and Models, AGU Fall meeting, Washington D.C., 2018. PDF
• Vesselinov, V.V., Novel Machine Learning Methods for Extraction of Features Characterizing Complex Datasets and Models, Recent Advances in Machine Learning and Computational Methods for Geoscience, Institute for Mathematics and its Applications, University of Minnesota, 2018. PDF
• O'Malley, D., Vesselinov, V.V., Alexandrov, B.S., Alexandrov, L.B., Nonnegative/binary matrix factorization with a D-Wave quantum annealer PDF
• Vesselinov, V.V., Alexandrov, B.A, Model-free Source Identification, AGU Fall Meeting, San Francisco, CA, 2014. PDF

Presentations are also available at slideshare.net

Videos

• Vesselinov, V.V., Novel Machine Learning Methods for Extraction of Features Characterizing Complex Datasets and Models, Recent Advances in Machine Learning and Computational Methods for Geoscience, Institute for Mathematics and its Applications, University of Minnesota, 2018.

Examples

• Blind Source Separation (i.e. Feature Extraction)
• Source Identification

TensorDecompositions

TensorDecompositions is a general framework for Unsupervised Machine Learning (ML) using Nonnegative Matrix/Tensor decomposition algorithms.

NMFk/NTFk (Nonnegative Matrix Factorization/Nonnegative Tensor Factorization) are two of the codes within the TensorDecompositions perform.

Unsupervised ML methods can be applied for feature extraction, blind source separation, model diagnostics, detection of disruptions and anomalies, image recognition, discovery of unknown dependencies and phenomena represented in datasets as well as development of physics and reduced-order models representing the data. A series of novel unsupervised ML methods based on matrix and tensor factorizations, called NMFk and NTFk have been developed allowing for objective, unbiased, data analyses to extract essential features hidden in data. The methodology is capable of identifying the unknown number of features charactering the analyzed datasets, as well as the spatial footprints and temporal signatures of the features in the explored domain.

TensorDecompositions algorithms are written in Julia.

NMFk/NTFk codes are available as open-source on GitHub

NMFk/NTFk use TensorFlow, PyTorch, MXNet, and MatLab

MADS

MADS (Model Analysis & Decision Support) MADS is an integrated open-source high-performance computational (HPC) framework.

MADS can execute a wide range of data- and model-based analyses:

• Sensitivity Analysis
• Parameter Estimation (PE), Model Inversion and Calibration
• Uncertainty Quantification (UQ)
• Model Selection and Model Averaging
• Model Reduction and Surrogate Modeling
• Machine Learning and Blind Source Separation
• Decision Analysis and Support

MADS has been tested to perform HPC simulations on a wide-range multi-processor clusters and parallel environments (Moab, Slurm, etc.).

MADS utilizes adaptive rules and techniques which allows the analyses to be performed with a minimum user input.

MADS provides a series of alternative algorithms to execute each type of data- and model-based analyses.

MADS can be externally coupled with any existing simulator through integrated modules that generate input files required by the simulator and parse output files generated by the simulator using a set of template and instruction files.

MADS also provides internally coupling with a series of built-in analytical simulators of groundwater flow and contaminant transport in aquifers.

MADS has been successfully applied to perform various model analyses related to environmental management of contamination sites. Examples include solutions of source identification problems, quantification of uncertainty, model calibration, and optimization of monitoring networks.

MADS current stable version has been actively updated.

Professional softwares/codes with somewhat similar but not equivalent capabilities are:

• DAKOTA
• UCODE
• PEST
• AGNI / ASCEM (under development)

MADS source code and example input/output files MADS website.

MADS documentation is available at github and gilab.

MADS old sites: LANL, LANL C, LANL Julia, LANL Python

WELLS

WELLS is a C code simulating drawdowns caused by multiple pumping/injecting wells using analytical solutions.

WELLS can represent pumping in confined, unconfined, and leaky aquifers.

WELLS applies the principle of superposition to account for transients in the pumping regime and multiple sources (pumping wells).

WELLS can apply a temporal trend of water-level change to account for non-pumping influences (e.g. recharge trend).

WELLS can account early time behavior by using exponential functions (transmissivities and storativities; Harp and Vesselinov, 2013).

WELLS analytical solutions include:

• confined aquifer (Theis, Mishra et al)
• unconfined aquifer (transformed Theis, Mishra & Neuman)
• leaky confined aquifer (Hantish, Mishra et al)
• leaky unconfined aquifer (Mishra et al)
• fully and partially penetrating pumping well(s)
• fully and partially penetrating observation well(s)
• transient pumping rates: step changes and linear changes (Mishra et al)

WELLS has been applied to decompose transient water-supply pumping influences in observed water levels at the LANL site (Harp and Vesselinov, 2010a). For example, the figure below shows simulated drawdowns caused by pumping of PM-2, PM-3, PM-4 and PM-5 on water levels observed at R-15.

Codes with similar capabilities are AquiferTest. AquiferWin32, Aqtesolv, MLU, and WTAQ.

WELLS source code, example input/output files, and a manual are available at the WELLS websites: LANL GitLab Julia GitHub

LA-CC-10-019, LA-CC-11-098

MPEST / MPRUN

MPEST is a LANL-develeoped parallel version of the code PEST (Doherty 2009).

MPEST has been developed to optimize the solving of parallel optimization problems at the LANL multi-processor clusters.

MPEST has been applied in many parallel computing projects worldwide.

MPEST parallelization framework is using the code MPRUN, a code that has been also developed at LANL.

MPRUN is using POSIX threads to fork individual model runs to processors; this approach has been demonstrated to be very effective and superior to using MPI calls in terms of computational performance.

MPEST/MPRUN parallel subroutines are currently imported and further developed in the code MADS. The source code, example input/output files, and a manual are available at MADS website.

Send me email

DiaMonD: Mathematics at the Interfaces of Data, Models, and Decisions

LANL PI: Velimir V Vesselinov

DiaMonD is a project funded by the U.S. Department of Energy Office of Science.

DiaMonD addresses Mathematics at the Interfaces of Data, Models, and Decisions.

DiaMonD involves researchers from Colorado State University, Florida State University, Los Alamos National Laboratory, Massachusetts Institute of Technology, Oak Ridge National Laboratory, University of Texas at Austin, and Stanford University.

DiaMonD web site.

LANL Environmental Management Programs

PI of "Vadose Zone and Regional Aquifer Flow and Transport Modeling": Velimir V Vesselinov

Los Alamos National Laboratory (LANL) is a complex site for environmental management. The site encompasses about 100 km2 (37 square miles) of terrain with 600 m (2,000 feet) of elevation change, and an average rainfall of less than 300-400 mm (12 to 16 inches) per year. The site is intersected by 14 major canyon systems. Ecosystems within the site range from riparian to high desert and boast over 2,000 archaeological sites, as well as endangered species habitats.  The surface and subsurface water flow discharges primarily along the Rio Grande to the east of LANL. The Rio Grande traverses the Española basin from north to south; several major municipalities use the river water downgradient from LANL for water supply (Santa Fe, Albuquerque, El Paso/Juarez).

Regional aquifer beneath LANL

The regional aquifer beneath LANL is a complex hydrogeological system.  The regional aquifer extends throughout the Española basin, and is an important source for municipal water supply for Santa Fe, Los Alamos, Española, LANL, and several Native-American Pueblos. The wells providing groundwater from this aquifer for Los Alamos and LANL are located within the LANL site and in close proximity to existing contamination sites. The regional aquifer is comprised of sediments and lavas with heterogeneous flow and transport properties. The general shape of the regional water table is predominantly controlled by the areas of regional recharge to the west (the flanks of the Sierra de los Valles and the Pajarito fault zone) and discharge to the east (the Rio Grande and the White Rock Canyon Springs). At more local scales, the structure of groundwater flow is also influenced by (1) local infiltration zones (e.g., beneath wet canyons); (2) heterogeneity and anisotropy in the aquifer properties; and (3) discharge zones (municipal water-supply wells and springs). The aquifer is also characterized by well-defined, vertical stratification which, in general, provides sufficient protection of the deep groundwater resources.

The vadose zone, between the ground surface and the top of the regional aquifer, is about 180-300 m (600-1000 ft) thick. The vadose zone is comprised of sediments and lavas with heterogeneous flow and transport properties. The variably-saturated flow and transport through the thick vadose zone occurs through pores and fractures, and is predominantly vertical with lateral deviations along perching zones. The groundwater velocities in the vadose zone are high beneath wet canyons (up to 1 m/a) and low beneath the mesas (1 mm/a). Due to complexities in local hydrogeologic conditions, the hydraulic separation between the regional aquifer and the vadose zone is difficult to identify at some localities, especially where mountain-front recharge is pronounced.

The complexity and size of the LANL site make environmental management a continuing engineering and scientific challenge. Legacy contamination—both chemical and radioactive—exists at many locations. Some of the oldest worldwide radioactive Material Disposal Areas (MDA’s), where waste is buried in pits and shafts, are located on the site. LANL is mandated to follow timetables and requirements specified by the Compliance Order on Consent from the New Mexico Environment Department (NMED) for investigation, monitoring, and remediation of hazardous constituents and contaminated sites. Currently, all the remediation activities are scheduled for completion in 2015.  LANL is taking actions to prevent potential contaminant effects on human health and the environment.

The environmental work performed at the LANL site is managed by the Environmental Programs (EP) Directorate. A team of external and LANL (Computational Earth Sciences Group, Earth & Environmental Sciences) researchers is tasked by the EP Directorate to provide modeling and decision support to enable scientifically-defensible mitigation of the risks associated with various LANL sites. The principal investigator of this team is Velimir Vesselinov.

Since the 1950's, the LANL site has been the subject of intensive studies for characterization of the site conditions, including regional geology and hydrogeology. Various types of research have been performed at the site related to contaminant transport in the environment which include (1) laboratory experiments, (2) field tests, and (3) conceptual and numerical model analyses. The work is presented in a series of technical reports and peer-reviewed publications.

Currently, important aspects of the environmental management at the LANL site include:

>
• design of a long-term monitoring network of groundwater flow and transport in the vadose zone and regional aquifer;
• investigation of the hexavalent chromium plume in the regional aquifer; and
• model-based analyses of the environmental impact caused by Material Disposal Areas (MDA’s): performance assessment (PA) and corrective measures evaluations (CME).

Chromium plume in the regional aquifer

A chromium plume has been identified in the regional aquifer beneath the LANL site. Our team has been tasked with providing modeling decision support to the Environmental Programs (EP) Directorate to enable scientifically-defensible mitigation of the risks associated with chromium migration in the environment. Large amount of data and information are available related to the chromium site (vadose-zone moisture content, aquifer water levels, contaminant concentrations, geologic observations, drilling logs, etc.); they are used to develop and refine conceptual and numerical models of the contaminant transport in the environment. The development of numerical models and performance of model analyses (model calibration, sensitivity analyses, parameter estimations, uncertainty quantification, source identification, data-worth analyses, monitoring-network design, etc.) is a computationally intensive effort due to large model domains, large numbers of computational nodes, complex flow media (porous and fracture flow), and long model-execution times. Due to complexities in the model-parameter space, most of the model analyses require a substantial number of model executions. To improve computational effectiveness, our team utilizes state-of-the-art parallel computational resources and novel theoretical and computational methods for model calibration, uncertainty analysis, risk assessment and decision support.

Numerical modeling of flow and transport in the regional aquifer near Sandia Canyon

The numerical model is capturing current conceptual understanding and calibrated against existing data (taking into account uncertainties)

Regardless of existing uncertainties, the model provide information related to:

>
• spatial distribution of contaminant mass,
• contaminant flux to the regional aquifer,
• monitoring-network design, and
• environmental risk.

Relevant computer codes:

• MADS: Model Analysis & Decision Support
• WELLS: Analytical simulator of drawdowns caused by multiple pumping wells

ASCEM

Advanced Simulation Capability for Environmental Management

Decision Support PI: Velimir V Vesselinov

A consortium of multiple national laboratories is developing high performance computer modeling capabilities to meet the challenge of waste disposal and cleanup left over from the creation of the US nuclear stockpile decades ago. The project is funded by the Department of EnergyOffice for Environmental Management (DOE-EM).

Within ASCEM, the goal of the "Decision Support" task is to create a computational framework that facilitates the decision making by site-application users, modelers, stakeholders, and decision/policy makers. The decision-support framework leverages on existing and novel theoretical methods and computational techniques to meet the general decision-making needs of DOE-EM as well as the particular site-specific needs of individual environmental management sites.

The decision-support framework can be applied to identify what kind of model analyses should be performed to mitigate the risk at a given environmental management site, and, if needed, support the design of data-acquisition campaigns, field experiments, monitoring networks, and remedial systems. Depending on the problem, decision-support framework utilizes various types of model analyses such as parameter estimation, sensitivity analysis, uncertainty quantification, risk assessment, experimental design, cost estimation, data-worth (value of information) analysis, etc.

Relevant computer codes:

• MADS: Model Analysis & Decision Support