Loading ...
Sorry, an error occurred while loading the content.

[ai-geostats] a questions of using GSTAT to do Gaussian Conditional Simulation!

Expand Messages
  • Feng Liu
    Hi, Dear List: I am trying to use Gstat to do Gaussian Conditional Simulation on my soil C content data. I got 3 questions as following: 1. If I use set
    Message 1 of 3 , Jul 7, 2004
    • 0 Attachment
      Hi, Dear List:
       
      I am trying to use Gstat to do Gaussian Conditional Simulation on my soil C content data. I got 3 questions as following:
       
      1. If I use "set nsim=100" to do 100 times simulation, does the simulations follow a single random path or they will follow 100 different random paths? What should I do if I want to let them follow different random paths.
       
      2. Is it possible for me to write all the results of 100 simulations to one single file? How?
       
      3. Gstat needs gnuplot to do the plot, but I could not get gnuplot compiled and start in windows, could you please tell me how to do it if you have experience on this.
       
      Thanks a lot
       
      Feng Liu
      Graduate Student
      Texas A&m University
       
    • Ernesto Jardim
      Hi, Why don t you use gstat within R ? At least the 2 last questions will be solved. Regards EJ ... * By using the ai-geostats mailing list you agree to follow
      Message 2 of 3 , Jul 7, 2004
      • 0 Attachment
        Hi,

        Why don't you use gstat within R ?

        At least the 2 last questions will be solved.

        Regards

        EJ

        On Wed, 2004-07-07 at 16:09, Feng Liu wrote:
        > st1\:* { BEHAVIOR: url(#default#ieooui) } --> -->
        > Hi, Dear List:
        >
        > I am trying to use Gstat to do Gaussian Conditional Simulation on my
        > soil C content data. I got 3 questions as following:
        >
        > 1. If I use "set nsim=100" to do 100 times simulation, does the
        > simulations follow a single random path or they will follow 100
        > different random paths? What should I do if I want to let them follow
        > different random paths.
        >
        > 2. Is it possible for me to write all the results of 100 simulations
        > to one single file? How?
        >
        > 3. Gstat needs gnuplot to do the plot, but I could not get gnuplot
        > compiled and start in windows, could you please tell me how to do it
        > if you have experience on this.
        >
        > Thanks a lot
        >
        > Feng Liu
        > Graduate Student
        > Texas A&m University
        >
      • Edzer J. Pebesma
        ... Single path; run the process with nsim=1; 100 times to get 100 different paths (and rename the output each time). ... Not with maps (maps are univariate),
        Message 3 of 3 , Jul 19, 2004
        • 0 Attachment


          Feng Liu wrote:
          Hi, Dear List:
           
          I am trying to use Gstat to do Gaussian Conditional Simulation on my soil C content data. I got 3 questions as following:
           
          1. If I use "set nsim=100" to do 100 times simulation, does the simulations follow a single random path or they will follow 100 different random paths? What should I do if I want to let them follow different random paths.
           
          Single path; run the process with nsim=1; 100 times to get 100 different paths
          (and rename the output each time).
          2. Is it possible for me to write all the results of 100 simulations to one single file? How?
           
          Not with maps (maps are univariate), you can when using ascii column files with
          prediction locations;
          3. Gstat needs gnuplot to do the plot, but I could not get gnuplot compiled and start in windows, could you please tell me how to do it if you have experience on this.
           
          Install gnuplot, and make sure that a binary called gnuplot is present in
          the search path; the gnuplot folks tend to call it gnupltw32.exe or something
          like that; see also the set gnuplot = 'xxx'; command in gstat files.

          Also, I couldn't agree more with Ernesto: I find myself using gstat inside
          R or S-Plus almost 100% of the time I spend using gstat.
          --
          Edzer
        Your message has been successfully submitted and would be delivered to recipients shortly.