>Hello to all-
>I am new to the list and this is my first post. I am currently working on
>a masters thesis in Geology. I have mapped a mineralized shear zone and
>collected over 400 soil samples, had them analyzed and have put these
>values into an Arc-info format. I have received some technical assistance
>in doing this, and the maps themselves are for the most part complete.
>I am now working with the data in order to 'statistically' evaluate the
>strength of relationships among some trace elements (As, Au, Ba, Co, Cu,
>Mo, Pb, Sb, and Zn). I am currently using SigmaStat 2.0 to calculate the
>Pearson Product Moment Correlation Coefficents for the elements.
>My questions are:
>Will i be able to digitize my geologic map and perform statistical
>analysis of the values within particular lithologies (in Arcinfo) ?
>and How difficult will this be?
>Are there any graphical means to represent the correlation coefficients i
>have calculated with Sigmastat?
>What other software packages can you suggest for analyzing such a
>Much thanks in advance-
Hello, Mark McCullars
I am very much interested in what you are doing, and I also have done some
work on statistical and GIS applications in environmental geochemistry
(trace elements). The following are some tips for you, based on my own
* Trace elements usually do not follow normal distribution. Before
statistical analysis, you had better transform the datasets (to follow
* Outlying values do not belong to the population of your samples, therefore
you should detect them, and treat them in some special way (delete or
replace). Outlying samples should also be detected.
* Geologial maps could be digitized, and then you can overlay your element
contents with the maps. Therefore, you could explain some phenomena based
on the spatial distribution maps. However, when you use geological maps,
you should be careful that your samples may not always follow the geological
feature (such as bedrock), because the geological maps are summarized (based
on their scales).
* You can also carry out some spatial analysis, such as geostatistics (study
the variograms in detail), spatial autocorrelation and fractal, etc.
* You should not expect too high on GIS, as it is only a tool for us.
Dr. Chaosheng Zhang
Division of Geochemistry
Geological Survey of Sweden, SGU
P.O. Box 670
S-751 28 Uppsalla
Tel: +46 18 179290
Fax: +46 18 179210
*To post a message to the list, send it to ai-geostats@...
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@...
with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!