Kalf and Associates Pty Ltd

Hydrogeological, Numerical modeling specialists

 

Home

 

About Us

 

Research

 

Saline Disposal

Sustainable Yield

Stochastic K Generation

 

Stochastic generation of hydraulic conductivity (K) fields using Value Noise based functions

 

                   
                           Value noise and related turbulence textures (256 x 256 cell model)
                           a) Gaussian type     b) Gaussian type with increased granularity factor
                           c) d) anisotropic without and with warping   e) f) simulated braided flood plain types

The aim of the research was to develop a Value-Noise based function approach for generation of stochastic hydraulic conductivity (K) fields for use in groundwater solute transport models.

The Value Noise based function is closely related to Perlin Noise and other similar functions used extensively for generating procedural textures in graphical images.

The complete method is based on creating a base cubic interpolated and smoothed “pseudo random” noise array which is then calculated for various scales combined by super position.  This super position is commonly referred to as “turbulence”.

A talk titled "Pre-Conditioned Monte Carlo simulation using a calibrated model and Value-Noise  stochastic K-fields" was presented at the 2nd Australasian hydrogeological research conference in Adelaide Australia 3 to 5 December 2006. A Powerpoint PDF file copy is available here Pre-Conditioned Monte Carlo simulation using a calibrated model and Value-Noise  stochastic K-fields (1.8Megs PDF)

I would be pleased to hear any comments. Contact me at frkalf@aol.com