Reservoir Modelling

Question 1

What issues are there with reservoir modelling?

Answer 1

How do you define the structure
What properties to put in each cell
BALANCE between detail and computability

Question 2

What are the 4 methods of building a static model?

Answer 2

Object modelling
Variograms
Multipoint statistic
Monte Carlo method

Question 3

What is a variogram?

Answer 3

Represents the spatial correlation of a variable
The simples means of describing spatial correlation between data points
Closely spaced samples are more likely to have a higher correlation than those spread further apart and that beyond a certain distance 0 correlation is reached
Problems with curvature

Question 4

What is multipoint statistics?

Answer 4

It is an alternative approach to categorical geostatistics that uses instead of a variogram, sample images to estimate the conditional probably at interpolation (a method of constructing new data points with the discrete set of known data point) location give the observed and the already interpolated data
FIELD APPLICATION - collection of training images, identification and characterisation of lithofacies, simulated of spatial variability, conditional of simulated to a well field
Cheap but not feasible fo collect 3D images in the field (need of sequential data)

Question 5

What is the Monte Carlo method?

Answer 5

Geological heterogeneity is typically complex to deal with using analytical statistical base on probability theory. Probability theories often to simple.
Run a model thousands of times to understand the result

Question 6

What is static reservoir modelling?

Answer 6

It is digital representation of the porosity and permeability distribution in the subsurface typically hydrocarbon bearing structure
THIS INCLUDES; tectonic features (faults and fractures) and as much sedimenntological features as possible. From sesimic, surface outcrop and well logs. Doesnt show fluvial motion