Application of a coupled adjoint sensitivity and kriging approach to calibrate a groundwater flow model

LaVenue, A.M.; Pickens, J. F.

doi:10.1029/92wr00208

Cited by 108 publications

(53 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…L'approche des points pilotes (Certes & De Marsily, 1991;Lavenue & Pickens, 1992;Lavenue et al, 1995;Ramarao et al, 1995) consiste en une méthode bayésienne et non linéaire. Les points pilotes sont utilisés dans les équations de krigeage comme des points ordinaires (Rentier, 2002), mais les valeurs sont des inconnues que l'on ajuste peu à peu pour caler le modèle.…”

Section: Modflow Pesunclassified

Modélisation directe et inverse de l'écoulement souterrain dans les milieux poreux

Toto

Zouhri

Jgounni

2009

Hydrological Sciences Journal

View full text Add to dashboard Cite

Résumé Le présent article montre l'intérêt des simulations mathématiques dans la compréhension des aquifères complexes et hétérogènes. Le réservoir côtier du Maroc occidental est un exemple qui traduit une répartition spatiale des corps perméables et imperméables, d'où la nécessité d'approfondir la connaissance de son fonctionnement hydrogéologique en termes de modélisation numérique. Le modèle hydrodynamique proposé repose sur la résolution du problème direct et inverse. La simulation de l'écoulement (problème direct) en régime permanent a montré la persistance de grands écarts entre la piézométrie mesurée et calculée. Cette anomalie de calage est expliquée principalement par le caractère hétérogène et fissuré, incompatible avec la dérivation de l'équation de diffusivité qui présuppose un milieu continu et un régime laminaire plutôt que turbulent. Deux cartes principales ont été obtenues: (a) une carte des polygones de la conductivité hydraulique dont les valeurs varient entre 50 10 -4 et 80 10 -4 m/s; et (b) une carte des polygones de recharge avec des valeurs comprises entre 4.5 10 -9 et 5 10 -14 m/s. Le calage du modèle d'écoulement recherché dans la résolution du problème inverse consiste à identifier les paramètres de l'aquifère à partir de la piézométrie connue. Le code d'inversion PEST (Parameter ESTimation) a été utilisé pour créer des cartes montrant des polygones de la conductivité hydraulique et de la recharge avec des valeurs très similaires aux résultats du modèle conceptuel.Mots clefs eaux souterraines; modélisation numérique; méthode directe; méthode inverse; code Groundwater Modelling System; PEST; points pilotes; caractéristiques hydrauliques; Maroc Direct and inverse modelling of the groundwater flow in porous mediaAbstract This article shows the role of mathematical simulation in the understanding of complex and heterogeneous aquifers. The coastal aquifer of the western Morocco is an example that reflects the spatial distribution of permeable and impermeable bodies. Hence the need to know its hydrological functioning in terms of numerical modelling. The hydrodynamic model proposed is based on solving the direct and inverse problem. In the steady state, the groundwater flow simulation (direct problem) showed the persistence of major differences between the measured and calculated heads. This calibration anomaly is mainly explained by heterogeneity and fissures, that are incompatible with the derivation of the diffusivity equation, which assumes a continuous and a laminar rather than turbulent flow. Two key maps have been obtained: (a) a map of hydraulic conductivity polygons whose values vary between 50 10 -4 and 80 10 -4 m/s; and (b) a map of recharge polygons whose values vary between 4.5 10 -9 and 5 10 -14 m/s. Calibration of the flow model, that was investigated in the solution of the inverse problem, leads to identification of the parameters of the aquifer from the known piezometric data. The inversion code PEST (Parameter ESTimation) was used to create maps of hydraulic conductivity and ...

show abstract

Section: Modflow Pesunclassified

Modélisation directe et inverse de l'écoulement souterrain dans les milieux poreux

Toto

Zouhri

Jgounni

2009

Hydrological Sciences Journal

View full text Add to dashboard Cite

show abstract

“…Randomized Maximum Liklihood (RML) 44,45 and Pilot Point 46,47,48 , attempt to be more efficient than the rigorous methods while providing realizations that approximate the target distribution. All methods require running the flow simulation which, for high resolution models, can be very expensive.…”

Section: Uncertainty Assessment Methodsmentioning

confidence: 99%

Streamline-Based Production Data Integration in Naturally Fractured Reservoirs

Al-Harbi

Cheng

et al. 2005

SPE Journal

View full text Add to dashboard Cite

Streamline-based models have shown great potential in reconciling high resolution geologic models to production data. In this work we extend the streamline-based production data integration technique to naturally fractured reservoirs. We use a dualporosity streamline model for fracture flow simulation by treating the fracture and matrix as separate continua that are connected through a transfer function. Next, we analytically compute the sensitivities that define the relationship between the reservoir properties and the production response in fractured reservoirs. Finally, production data integration is carried out via the Generalized Travel Time inversion (GTT). We also apply the streamline-derived sensitivities in conjunction with a dual porosity finite difference simulator to combine the efficiency of the streamline approach with the versatility of the finite difference approach. This significantly broadens the applicability of the streamlinebased approach in terms of incorporating compressibility effects and complex physics.The number of reservoir parameters to be estimated is commonly orders of magnitude larger than the observation data, leading to non-uniqueness and uncertainty in reservoir parameter estimate. Such uncertainty is passed to reservoir response forecast which needs to be quantified in economic and operational risk analysis. In this work we sample parameter uncertainty using a new two-stage Markov Chain Monte Carlo (MCMC) that is very fast and overcomes much of its current limitations. The computational efficiency comes through a substantial increase in the acceptance rate during MCMC by using a fast linearized approximation to the flow simulation and the likelihood function, the critical link between the reservoir model and production data. The Gradual Deformation Method (GDM) provides a useful framework to preserve geologic structure. Current dynamic data integration methods using GDM are inefficient due to the use of numerical sensitivity calculations which limits the method to deforming two or three models at a time. In this work, we derived streamline-based analytical sensitivities for the GDM that can be obtained from a single simulation run for any number of basis models. The new Generalized Travel Time GDM (GTT-GDM) is highly efficient and achieved a performance close to regular GTT inversion while preserving the geologic structure.

show abstract

“…The result is a smooth variation of the hydraulic property over the model domain. Numerous studies have used pilot points for groundwater model calibration (de Marsily and others, 1984;LaVenue and Pickens, 1992;Petkewich and Campbell, 2007) and have proven them to be powerful, flexible tools for representing spatial heterogeneity in various types of aquifer-hydraulic properties.…”

Section: Calibration Proceduresmentioning

confidence: 99%

Numerical simulation of the groundwater-flow system of the Kitsap Peninsula, west-central Washington

Frans¹,

Olsen²

2016

Scientific Investigations Report

View full text Add to dashboard Cite

For more information on the USGS-the Federal source for science about the Earth, its natural and living resources, natural hazards, and the environment-visit http://www.usgs.gov or call 1-888-ASK-USGS.For an overview of USGS information products, including maps, imagery, and publications, visit http://store.usgs.gov.Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.Although this information product, for the most part, is in the public domain, it also may contain copyrighted materials as noted in the text. Permission to reproduce copyrighted items must be secured from the copyright owner. DatumsVertical coordinate information is referenced to the North American Vertical Datum of 1988 (NAVD 88).Horizontal coordinate information is referenced to the North American Datum of 1983 (NAD 83).Altitude, as used in this report, refers to distance above the vertical datum. Simulated inflow to the model area for the 2005-2012 period from precipitation and secondary recharge was 585,323 acre-feet per year (acre-ft/yr) (93 percent of total simulated inflow ignoring changes in storage), and simulated inflow from stream and lake leakage was 43,905 acre-ft/yr (7 percent of total simulated inflow). Simulated outflow from the model primarily was through discharge to streams, lakes, springs, seeps, and Puget Sound (594,595 acre-ft/yr; 95 percent of total simulated outflow excluding changes in storage) and through withdrawals from wells (30,761 acre-ft/yr; 5 percent of total simulated outflow excluding changes in storage). AbbreviationsSix scenarios were formulated with input from project stakeholders and were simulated using the calibrated model to provide representative examples of how the model could be used to evaluate the effects on water levels and stream baseflows of potential changes in groundwater withdrawals, in consumptive use, and in recharge. These included simulations of a steady-state system, no-pumping and return flows, 15-percent increase in current withdrawals in all wells, 80-percent decrease in outdoor water to simulate effects of conservation efforts, 15-percent decrease in recharge from precipitation to simulate a drought, and particle tracking to determine flow paths.Changes in water-level altitudes and baseflow amounts vary depending on the stress applied to the system in these various scenarios. Reducing recharge by 15 percent between 2005 and 2012 had the largest effect, with water-level altitudes declining throughout the model domain and baseflow amounts decreasing by as much as 18 percent compared to baseline conditions. Changes in pumping volumes had a smaller effect on the model. Removing all pumping and resulting return flows caused increased water-level altitudes in many areas and increased baseflow amounts of between 1 and 3 percent.

show abstract

Application of a coupled adjoint sensitivity and kriging approach to calibrate a groundwater flow model

Cited by 108 publications

References 50 publications

Modélisation directe et inverse de l'écoulement souterrain dans les milieux poreux

Modélisation directe et inverse de l'écoulement souterrain dans les milieux poreux

Streamline-Based Production Data Integration in Naturally Fractured Reservoirs

Numerical simulation of the groundwater-flow system of the Kitsap Peninsula, west-central Washington

Contact Info

Product

Resources

About