What are the software packages for geostatistical analysis, in addition to ArcGIS Geostatistical Analyst (paid and free)? And what are the software packages for fuzzy geostatistical analysis?
I have experiences with it during my doctorate and it is of acceptable quality and hence can be recommended.However it is running under DOS. But i think it should be possible to get it running in a virtual machine with the free software VM ware player. when you had installed a DOS before.
Then i have seen, that variogram analysis also seems to be possible with R. Also Kriging interpolation, Spatial BLUP (best linear unbiased prediction) is possible with R. I have seen, that the packages geoR and gstat and a package Kriging are available with R. Usually routines, written for R shold have a good quality, so also this should give you an opportunity.
I have experiences with it during my doctorate and it is of acceptable quality and hence can be recommended.However it is running under DOS. But i think it should be possible to get it running in a virtual machine with the free software VM ware player. when you had installed a DOS before.
Then i have seen, that variogram analysis also seems to be possible with R. Also Kriging interpolation, Spatial BLUP (best linear unbiased prediction) is possible with R. I have seen, that the packages geoR and gstat and a package Kriging are available with R. Usually routines, written for R shold have a good quality, so also this should give you an opportunity.
I have experiences with it during my doctorate and it is of acceptable quality and hence can be recommended.However it is running under DOS. But i think it should be possible to get it running in a virtual machine with the free software VM ware player. when you had installed a DOS before.
Then i have seen, that variogram analysis also seems to be possible with R. Also Kriging interpolation, Spatial BLUP (best linear unbiased prediction) is possible with R. I have seen, that the packages geoR and gstat and a package Kriging are available with R. Usually routines, written for R shold have a good quality, so also this should give you an opportunity.
The GIS package ArcGIS for Desktop with the extension Geostatistical Analyst provide geostatistical analysis (www.esri.com). The co-creator of the extension, Krivoruchko K, authored a tutorial titled Spatial statistical data analysis for GIS users which gives good advice how to use Geostatistical Analyst. The package Surfer V13 (www.goldensoftware.com/products/surfer) contains support functions for geostatistical analysis.
You could try ISATIS of Geovariances (http://www.geovariances.com/en/isatis-all-in-one-software-for-geostatistics-ru324) or SGemS (http://sgems.sourceforge.net/) an open-source software. Good luck!
Several commercial packages have been mentioned (ISATIS,, Geostatisical analyst with ArcGIS, Sufer) but the question might first be, how knowledgeable are you about geostatistics? As with any statistics package you can't expect to just dump your data in, run the program and expect to get useful results. You the user have to make a number of decisions and also ask whether certain statistical assumptions are satisfied. Probably the easiest package to use is called GEO-EAS (free from EPA, it precedes GEOPACK). As was noted above it was written to function with DOS but it is very interactive, it comes with not only a manual but also tutorial material. It has quite good graphics. There are many components in this package and all but a few can handle quite large data sets.
For an example/tutorial using this package see
.Interpolation and Estimation with Spatially Located Data Chemometrics and Intelligent Laboratory Systems 11, (1991) 209-228.
Before attempting to use specific geostatistical tools (e.g., estimating & modeling a variogram, point kriging, block kriging, simulation) you will want to use various exploratory tools (histograms of the data values, coded plots of the data locations, trend surface plots of data values vs polynomial functions of the location coordinates) and have a reasonable good understanding of the phenomenon that generates the data. the difference between "point" and "non-point" data is very important. Particularly in mining applications you are likely to want to use point data to estimate block values.
Both gstat and GeoR (R libraries) are very good, each has some features that the other does not. While R is not hard to learn you need to something about it to get your data into the programs.
For Geostatistical Analyst you need to know how to use ARCGIS, the learning curve for ISATIS is much steeper and quite a bit more expensive, it was written for workstations. If you want to use the geostatistics software with a GIS package then do a search for R and GIS on Google, there are several open source GIS packages
If your data is multivariate then you need other exploratory statistics tools, see XGOBI for one, it is open source and there is good tutorial material available on the internet and with the software (it works well with R)
There are some geostatistical tools in SAS but you need to know SAS first and the geostatistics components look quite different from the software mentioned above
Indeed ArcGIS is maybe one of the most famous and comprehensive commercial GIS packages which allows Geostatistical analysis, between other applications. Have a look on this: http://resources.arcgis.com/en/help/main/10.1/index.html#//003100000004000000
At any rate, there are several powerful open source packages such as QGIS or SAGA-GIS as underlined by others answers.
Plenty of information and tutorials do exist about the aforementioned packages.
From set of commercial packages ArcGIS module Geostatistical Analysis is the best. If you look on the Open Source packeges I suggest : GRASS, QGIS and SAGA GIS.
There is also R programming ,ADaMSoft – a generalized statistical software with data mining algorithms and methods for data management. These are open source software. The choice also depends on the objective of your analysis and the data volume
You can try also the GeoMS2 package from the Cerena research group of Lisbon (Prof. Amilcar Soares and collaborators). Includes basic statistics, structural analysis -variogram-, estimation and simulation processes. GeoMS2 is a Phyton application with a very good graphic interface.
I will come to you and additively to my answer yesterday, it is desirable to identify the nature you realize constantly and from this, it should help you to choose objectively the best software for your applications.
R is the best for statistics and geostatistics in general. It interfaces with all the Open Source GIS and there's a package almost for everything, just like, for example, the Fuzzy Set Ordination: http://ecology.msu.montana.edu/labdsv/R/labs/lab11/lab11.html or Fuzzy Analysis Clustering: https://stat.ethz.ch/R-manual/R-devel/library/cluster/html/fanny.html ..I don't know what kind of fuzzy analysis you have to do specifically BUT you can ask Google your necessity by typing e.g. "Fuzzy k-means analysis" + "R cran" and you will find, for sure the R package which is right for you.
R is the best for statistics and geostatistics in general. It interfaces with all the Open Source GIS and there's a package almost for everything, just like, for example, the Fuzzy Set Ordination: http://ecology.msu.montana.edu/labdsv/R/labs/lab11/lab11.html or Fuzzy Analysis Clustering: https://stat.ethz.ch/R-manual/R-devel/library/cluster/html/fanny.html ..I don't know what kind of fuzzy analysis you have to do specifically BUT you can ask Google your necessity by typing e.g. "Fuzzy k-means analysis" + "R cran" and you will find, for sure the R package which is right for you.
Best,
Annalisa
ps. otherwise you can have a look at all the fuzzy possibilities here: https://cran.rstudio.com/web/packages/available_packages_by_date.html
GeoDa, QGIS, SAGA GIS and GRASS are not geostatistics software packages. Several interface with R and hence you would have access to some geostatistics packages such as gstat (R is really more like a programming language). For ARCGIS you have to have the geospatial analyst add-on (extra cost).
It depends somewhat on the kind of data you have, note that GRASS is raster based whereas ARCGIS is vector based (shape files). There is a package in R to read shape files. For very large data sets you may want to use SQL with the geostatistics software, R makes this possibe
It also depends on how much you already know about geostatistics and one or more of the software packages mentioned above, i.e. some have steeper learning curves than others. For example there are some "procs" in SAS that will allow you to do geostatistics but the learning curve is steeper than say R and gstat (SAS is commercial). ISATIS is another commercial software package, actually moderately expensive.
If you want a free and really easy geostatistics software package do a search on Google for GeoEAS, it is actually written for DOS but has quite good graphics. You can get both binaries and source code for free as well as a good user manual (all from US EPA). There are some limitations on the size of the data sets.
Here is a tutorial on the use of GEOEAS
1991, Myers,D.E.Interpolation and Estimation with Spatially Located Data Chemometrics and Intelligent Laboratory Systems 11, 209-228.
Once you have learned to use this software, more sophisticated software will be easier
Dear collegues, my university has license ArcGIS 8, 9 including ArcGIS Geostatistical Analyst (only 1 Workplace unfortunately). I work in ArcGIS (only licensed) more 10 years. But I search the open soft because it's soft is more spread. Besides that our university not buy a new version of ArcGIS. I study R (finished some MOOC on Coursera in USA universities). I am interest a new expierience.
Best software for geostatistical simulation? without any doubt sgems developped at Stanford (you can find it on source forge.net). It is free, up to date and has its own 3D visualization interface.
Whenever you want to label something as "best" you need to identify and disclose the criteria for making the judgement. In the case of software there are always conflicting criteria, e.g. cost, availability of source code, platforms for binaries, tutorial/documentation availability, 32 bit vs 64 bit, ease of use, learning curve, graphics, full disclosure of algorithms used, size of allowed data sets, 2-D vs 3-D, linkage to other and usage with other software, evidence that the software is reliable, provisions for maintaining the software (fixing bugs, adding updates and/or new features, support), breadth of user population (both in terms of numbers and also across disciplines), ease of establishing reproducibility of results, is it general purpose software or specialized, etc (not intended to be a complete list).
The variation in the software packages recommended above likely is a consequence of the differences in criteria used and the relative importance assigned to various criteria. In sum there really isn't a "best" choice because the individual user will have their own set of criteria and/or constraints. There is also the question of who is making the decision to obtain/install/maintain the software (one individual, a small group or a big organization). Does the software provider accept any responsibility for erroneous results and consequences of erroneous results?
Erwan, the public version of SGeMS is not maintained in years. It is incorrect to say it is up to date. Old releases are hosted on sourceforge as you mentioned.