Science method

Geomodelling - Science method

Explore the latest questions and answers in Geomodelling, and find Geomodelling experts.
Questions related to Geomodelling
  • asked a question related to Geomodelling
Question
1 answer
We have got quite old version of GModeller - perhaps the problem is explained or resolved already.
Currently we are working on 3D model of Suwalki Anorthosite Massif. Horizontal size of the model is 50 x 90km. Even if we try to calculate forward gravity model only for crystalline basement, the answer is dominated by strong aliasing at the southern border of modelled area. It dissapears when we tried resolution 5x5 km horizontally and 10m verticaly, but such resolution is of course far too low!
Are there other hints or tips to get rid of the aliasing?
Relevant answer
Dear Olga,
I can give you a conceptual answer such that you can your self define your sampling frequency.
The criterion defining the sampling frequency fsx in a certain direction say x, is that one has to be able to reconstruct again the original waveform f(x).
Nyquist gave an answer to this problem such that one has to sample f(x) such that fsx>= fmx, where fmx is the maximum contained frequency in the waveform f(x). Then all what you need is to make Fourier transform of your waveform to get F(w) where w is the angular frequency.
If you want to prevent aliasing above retrain frequency you have to pass your signal at first through antialiasing filter to limit the highest frequency fxm of your f(x) to fsx/2.
This is my advise to you. You can use an anti aliasing filter to limit the bandwidth of your signal to fsx/2
Best wishes
  • asked a question related to Geomodelling
Question
5 answers
Dear Colleagues,
As one of the researchers in ANN applications in air pollution field, I suggest you or your team to make the ANN-data to a real application based on ANN model, moreover it can be a dynamic model. If you think, it is possible make this project together, indeed, I guess that such a study will fill a gap in this modeling area for air pollution, control and understanding. I will be happy to get your answer about this start-up and use data in ANN-modeling to make a real, executable and dynamic application. Currently, I've developed one ANN-based PM10 model but it should be enhanced and actually not a dynamic one. Please, dont hesitate to contact me.
Dr. Fatih Taşpınar
Relevant answer
Prof. Faith
I am studying the vertical dispersion of air pollutants for a domain of 1km x 1km.
We have models only to predict vertical dispersion in street canyon and horizontal dispersion model to predict concentration at a specific distance from the source.
I wish to analyse the vertical dispersion of pollutants in a urban region.
Also to study the influence of meteorological parameters on pollutant dispersion and from this simulation i could arrive with eddy hot spots and turbulence.
  • asked a question related to Geomodelling
Question
9 answers
I am trying to find a solution for the thermal conduction equation:
d ( K[T(z)] * d T(z) ) = -H0* e (-z/d)
dz                dz
Some publications have proposed methods based on stochastic models and nonlinear transformations. However, I would like to know if there is another way to estimate continental geotherms considering the temperature-dependent thermal conductivity. I am following the next relation for the thermal conductivity of the crust:
K [T(z)] = 2.26 - 618.241/  T(z) +K0 * (255.576/  T(z) - 0.030247)
                                              
I will appreciate any help.
Relevant answer
Answer
In essence, your equation takes the form,
[ (a + b/T)T' ]' = - exp(-z)
where a and b are constants and I have missed a few constants out just for this "proof of concept" idea. Integrate once to get
[a + b/T] T' = exp(-z) + c      where c=constant of integration.
This latest equations is of variables-separable form, and therefore its integral is
aT + b lnT  =  -exp(-z) + cz + d   where d=constant of integration.
This can't be rearranged to write T explicitly interms of z (or vice versa) but if we know c and d then a straightforward Newton-Raphson scheme will find T as z varies.  If the necessary boundary conditions are given at one value of z, then c and d may be found. LIkewise if T is given at two different values of z. It might be a little more tricky if T and T' are given at different locations.
  • asked a question related to Geomodelling
Question
4 answers
I am researching the thermal effects of a culvert embedded in the ground in an area of permafrost. We have installed the 3 different culverts, all monitored by thermistor beads, which relay data back to us. I am looking for a program which will allow me to create a scatter plot, showing the thermistor beads, as well as a 3D cylindrical object, showing the culvert(s).
Thanks, 
Shea Smith.
Relevant answer
Answer
Dear Shea,
FLAC3D software will be a good alternative to choose. 
Regards.
  • asked a question related to Geomodelling
Question
2 answers
I need to simulation a single well chemical tracer test (SWCTT) for surfactant polymer flooding of an oil well. the simulation results may next be matched with the field measurements. after the SWCTT simulation, a full field simulation will be performed based on the matching results.
the problem includes many physical aspects : multiphase porous media flow, non isothermal flow, surfactant and polymer odsorbtion....
we also are looking if we can load a heterogeneous geological model (space changing porosity and permeability) built in PETREL or other geomodeling software.
Relevant answer
Answer
Not sure about COMSOL. In principle in COMSOL you can define any system of partial differential equations that can solve a wide variety of problems, so I would be very surprised if it was not possible to solve the set of PDEs that describe surfactant-polymer flooding. You need Darcy type of flow, fractional flow with relative permeability etc.
However, there might be an easier solution. There is a publicly available simulator, UTChem
that has been specially designed for exactly this application.
So in principle UTChem is ready to use, you don't have to define any equations, the relevant ones are already built-in.
There is a small difference on the solver side. COMSOL uses the finite element method, I believe that UTChem is using finite volume, because finite volume is known to be very good in mass conservation.
  • asked a question related to Geomodelling
Question
4 answers
Department of Mining at Tallinn University of Technology is preparing proposals for Horizon 2020 calls and is ready to be partner also in other mining related proposals.
Department of Mining prepares applications for following calls:
SC5-11d-2015 - EC link: Topic: New sustainable exploration technologies and geomodels
Department of Mining is looking forward to be partner in following or related calls:
SC5-13c-2015 - EC link: Topic: Innovation friendly minerals policy framework
SC5-13e-2015 - EC link: Topic: Raw materials intelligence capacity
SC5-13f-2015 - EC link: Topic: Strategic international dialogues and cooperation with raw materials producing countries and industry
EO-2-2015 - EC link: Topic: Stimulating wider research use of Copernicus Sentinel Data
Department of Mining prepares applications for following calls under Central Baltic and Baltic Sea Programme:
CB - More exports by the Central Baltic companies to new markets and other Mining related priorities
BSR - Efficient management of natural resources and other Mining related priorities
Relevant answer
Answer
Hi Veiko,
I am interested in  SC5-13f-2015 - EC link: Topic: Strategic international dialogues and cooperation with raw materials producing countries and industry.
Hope you are well and greetings,
Vladimir
  • asked a question related to Geomodelling
Question
7 answers
I am working on my first advanced subsurface temperature model, that includes a lot of different model set ups, parameter sets and sensitivity analysis. In your opinion or from your experience, what procedure would you recommend how to document the evolution and the results of numerical models in order that model set up and results of the different model versions are still understandable in future (e.g. years after project end)?
Thanks in advance.
Relevant answer
Answer
"git" is the version control used within linux.  Registering your project on the github.com website will provide lots of extras features, including a hosted website for the project.
Beyond that, you should write a manual that will tell people how to use the model.  Be sure to include test cases and expected results.
In the expected results, it can be helpful to include numerical values to 10 or more digits, as a check that the code had not changed.  (These digits will not be dynamically significant but can help to see if someone edited the code and accidentally transposed digits, etc.)
Modern software normally comes with a test suite that runs examples and checks values numerically.
Use a Makefile to compile and drive the model.  This makes reproduceability much better.
Try to make the code immutable, with changes being only in user-generated driver files  or in commandline arguments.  This is very important to someone trusting the code.  For example, on my machine (OSX running the not-yet-released Yosemite operating system) I get as follows, which assures me that my file-listing command was written long ago.  Now, of course, the Apple programmers *could* have messed with the source code, but they are likely to agree that immutable code is a goal, and I would bet lunch money that they left that decade-old code alone.
$ strings `which /bin/ls` | head -4
$FreeBSD: src/bin/ls/cmp.c,v 1.12 2002/06/30 05:13:54 obrien Exp $
@(#) Copyright (c) 1989, 1993, 1994
The Regents of the University of California. All rights reserved.
$FreeBSD: src/bin/ls/ls.c,v 1.66 2002/09/21 01:28:36 wollman Exp $
  • asked a question related to Geomodelling
Question
5 answers
-
Relevant answer
Answer
Do you want to use your geochemical data (major- and trace elements) in order to find out whether or not you're dealing with FC-AFC-FCA or magma mixing processes? If so i think Ersoy & Helvaci (2010) published a paper that might be helpful, since it also contains an excel spreadsheet in the supplementary material.