Contents
How is a variogram used in spatial simulation?
Variogram is a measure of dissimilarity over a distance. It shows how two data points are correlated from a spatial perspective, and provides useful insights when trying to estimate the value of an unknown location using collected sample data from other locations.
How are data points separated in a variogram?
The equation for variogram observes the difference between these two data points: N ( h) accounts for all data point pairs that are separated by lag distance h. Although only horizontal separation is shown in the image, separation between two data points can be horizontal, vertical, or diagonal.
What happens when the H of a variogram increases?
As the h increases, there will be fewer number of pairs that are separated by h due to spatial limitation. Sill ( σ 2) is the variance in which spatial data pairs lose correlation. As the distance between two data points increases, it will be less likely that those two data points are related to one another.
What does low variance mean in a variogram?
In variogram, low variance ( γ) represents stronger correlation among data pairs. Spatial data pairs lose correlation with one another when variance ( γ) at lag distance ( h) reaches the sill ( σ 2 ). If there are variance poitns that exceeds the sill, it indicates the presence of trend, and needs to be detrended before variogram modeling.
Can a neutral isotropic model be constructed in Python?
According to an article written by Jared Deutsch, in the absence of strong geologic evidence, a neutral isotropic model can be constructed to assist in determining a principle direction (in later code implementation, isotropy is established by setting azi_tol=90. You are allowing for ± 90 degrees of azimuth tolerance).
What should the azimuth tolerance be in Python?
Azimuth tolerance should be chosen such that maximum number of pairs are found in the search template and exclude unreasonable associations. A common choice of azimuth tolerance is 22.5°. However, it can be changed to make a decision about whether to increase precision, or stability.
What kind of virtual environment does Python use?
The code snippet assumes Anaconda 5.2.0 version of Python virtual environment I would like to acknowledge Micahel Pyrcz, Associate Professor at the University of Texas at Austin in the Petroleum and Geosystems Engineering, for developing course materials that helped me write this article.