Ninth International Geostatistics Congress, Oslo, Norway
June 11 – 15, 2012
 
 
 
 
 
 
 

Session:

Environment 1

Abstract No.:

O-013

Title:

Construction of high-resolution stochastic geological models and optimal upscaling to MODFLOW

Author(s):

P. Quental, CICEGe FCT Universidade Nova Lisboa (PT)
J.A. Almeida, CICEGe FCT Universidade Nova Lisboa (PT)
M. Simões, CICEGe FCT Universidade Nova Lisboa (PT)
H. Amaral, LNEG (PT)
J. Fernandes, LNEG (PT)

Abstract:

Geostatistical geological models are good candidates for modelling of groundwater in a flow simulator, particularly in vertical delineation of aquifers and local allocation of hydraulic properties. However, the interface between geostatistical and flow simulator models is constrained by the spatial resolution. While several millions of blocks can define a stochastic geological model, dynamic simulation models are limited to tens of thousands of blocks.

The main objective of this work is the implementation of a methodology that optimises integration between high-resolution stochastic geological models and groundwater simulation models. The solution presented focuses on achieving a high-resolution geological model, taking advantage of geostatistical algorithms and subsequent optimal upscaling for a larger mesh.

The proposed methodology is divided into two main steps. In the first, a 3D stochastic geological model of the lithologies (high-resolution geological model) was constructed using sequential indicator simulation (SIS). The SSI algorithm included simultaneously two constraints, which is one of the innovations of this work: i) correction for local means, respecting the geological complexity of the area and the vertical occurrence of lithogroups; and ii) a histogram of vertical transitions between lithogroups.

In the second step, the geological model was upscaled / simplified according to an optimization procedure. To this end, first we defined the main hydrogeological units conceptually. Then, the option of simplifying the optimal stochastic geological model was carried out through an innovative software application developed, programmed and tested by us, based on the optimisation method simulated annealing. Limits are obtained for each of the hydrogeologic units, and an array of hydraulic parameters, that can be used directly in the flow model.

The proposed methodology was implemented for the underground area of the SPEL, in Seixal municipality. In addition to developing geological and hydrogeological models, synthetic tests of groundwater flow in MODFLOW were carried out.

   

 

 


    Copyright 2012 International Geostatistics Congress
 

Produced by X-CD Technologies Inc.