Really this is a complicated question in that there are rather a lot of variables and things to look out for.

I would sum up the main concerns to consist of interpreter experience (have you done this before etc...), the image data used (what imagery are you using, geology maps already published), and the guidelines (scale, type of map etc) laid out.

Data

Before you start have you looked at using something like OneGeology WMS feed to better inform the location you are looking at. This will include information that has already been mapped by national geological agencies.

OneGeology Portal: http://portal.onegeology.org/OnegeologyGlobal/

Furthermore, in what climatic environment is the investigated area being studied? (i.e. Tropical Jungle vs Desert environments). In this regard I would look to see what has been done before see how pronounced features have been interpreted and categorise from there.

Experience

If you will go down the raw interpreting satellite imagery (not radar) to derive faults, volcanic emission centres and structural lineaments. Things to be aware of including the very nature of geological interpretation being partly subjective and therefore results should be confirmed with field experience however, if this can't be done then you should just bear in mind experience in geological interpretation is important...

Have a look at this paper which might help: http://www.gsm.org.my/products/702001-100335-PDF.pdf

Dear Simon Kettle,

thank you very much for your answer. I have some papers recent published, for a general view of my area of study.

What I want to do is to deepen these features at a bigger and more detalied scale. I also have some aerial photographs.

Dear Marco,

You can easily detect faults or any other geological structural features using DEM derived shaded relief map. More accurate results come for higher resolution data-sets. However, extracting lineaments from google earth is not recommended cause many regions are characterized by dense vegetation or some other land-use pattern (Such as agriculture).

Accurate and reliable extraction is easier through terrain model. The disadvantage of extraction of lineaments using these data sets is that you can not discriminate which are truly geological structures and which are some other features (eg. geomorphic lineaments, man-made canals etc.). Therefore, a very extensive fieldwork is require to identify the geological features.

Here, I've attached my article recently published in Arabian Journal of Geosciences, maybe help you for your study.

Cheers.

Sumit

Thank you very much for your answer Sumit!

I will work with a good DEM map...

Cheers,

Marco

Dear Macro,

In my study I define faults from lineaments by their collinearity.

Faults zones contain multiple collinear lineaments.

Lemme recommend you my LEFA software (works under Matlab):

http://lefa.geologov.net

It has pdf guide:

Presentation LEFA (Lineament Extraction Fracture Analysis) v. 1.0 Analyti...

It can process both Landsat and SRTM.

The results can be saved in ESRI Shape format, than you can easily overlay resulting SHP over your data.

Regards, Sergei Shevyrev

Article CENTRAL PART OF THE OKHOTSK SEA BOTTOM RELIEF (VIEWING FROM SPACE)

Our modest experience with Google Earth and Google maps.

Sincerely