Skip to content

Latest commit

 

History

History
88 lines (61 loc) · 3.13 KB

README.md

File metadata and controls

88 lines (61 loc) · 3.13 KB
Raw

Falls Lake Agent Based Model

Dependencies

Install VineCopula from an R terminal. Type the following command:

install.packages("VineCopula")

VineCopula is a large package and will take some time to compile and install.

Coupling R and Java:

i.e. getting VineCopulas working:

In src/pertubing/GenerateTimeseries.java, update lines 20-33 depending on the OS used. The code committed here should work on Linux so long as Rscript is found in /usr/bin/. To find the location of Rscript, use:

which Rscript

in the terminal. Replace line 25 with whatever is output by the above command. On Windows, the forward slashes (/) should be replaced with double backslash (\\)

For Linux/ maybe Mac OSX:

String floc = System.getProperty("user.dir");
System.out.println(floc + "/analysis_flows.R ");
try {
  Process child =
      Runtime.getRuntime()
          .exec(
              "/usr/bin/Rscript "
                  + floc
                  + "/analysis_flows.R "
                  + Long.toString(seed)
                  + " "
                  + manage
                  + " "
                  + shifacStr
                  + " "
                  + Integer.toString(RunNum));

For Windows:

If using this program on Windows, overwrite lines 20-35 with the following code. Change line 26 to match your path location of Rscript.exe.

String floc = System.getProperty("user.dir");
System.out.println(floc + "\\analysis_flows.R ");
try {
  Process child =
      Runtime.getRuntime()
          .exec(
              "C:\\R\\R-3.6.1\\bin\\Rscript "
                  + floc
                  + "\\analysis_flows.R "
                  + Long.toString(seed)
                  + " "
                  + manage
                  + " "
                  + shifacStr
                  + " "
                  + Integer.toString(RunNum));

Running the ABM:

The starting parameters for the ABM can be found on lines 607-609 in WRRSim.java.

numOfShifts = 8; // 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.9, 1.0
numOfRun = 1;
numOfManagementScenarios = 1;
  • numOfShifts is converted to the target shift factor.
  • numOfRun is how many runs of each scenario will be simulated.
  • numOfManagementScenarios controls which mgmt scenarios will be run.

To run only 1 shift factor or 1 management scenario, be sure to change the value of m or n on lines 630-631 to be 1 less than the specified shift factor or management scenario.

Using the above parameters with m = 7 and n = 0 will produce a single model simulation with a shift factor of 0.7 and the first management scenario.

Output files:

  • probs-mgmt-x-s-x.x-x.csv - These are the probabilities generated by the Copulas
  • ts-mgmt-x-s-x.x-x.csv - These are the values of the climate variables produces by the Stochastic Reconstruction Framework
  • res-mgmt-x-s-x.x-x.csv - These are the ABM results.