RBBGCMuso/docs/CIRM

This is a si

Preparations

Before using this script, make sure, your current working directory looks like

.
├── glue.R
├── kichen_sink.R
├── make_individual_trees.R
├── Martonvasar_maize_KSH_Fejer.obs
├── Martonvasar_maize.obs
├── README.md
├── statistics.R
└── tree_accuracy.R

Loading the RBBGCMuso package and the necessary functions

library(RBBGCMuso)
source("make_individual_trees.R") # The DT creation and update algorithms
source("glue.R") # GLUE optimizer algorithms

The file containing the path to the observation files (Martonvasar_maize.obs), and the parameter intervals (Martonj)

Reading the observations

The mean yield had to be adjust. see in art.

measureFile <- "Martonvasar_maize.obs"
measurements <- read.csv2(measureFile, stringsAsFactors=FALSE)
measurements$mean <- measurements$mean / 10000
measurements$sd <- measurements$sd / 10000

Define conditioning functions

constraints.json

{
    "constraints": [

        {
            "Expression": "SELECT(harvest_index, max)|median",
            "Min": 0.45,
            "Max": 0.55
        },
        {
            "Expression": "SELECT(proj_lai, max)|quantile(.,0.5)",
            "Min": 2.7,
            "Max": 5
        },
        {
            "Expression": "SELECT(rootdepth5, max)|quantile(.,0.5)",
            "Min": 1.40,
            "Max": 1.80
        },			
        {
            "Expression": "SELECT(flower_date, max)|quantile(.,0.5)",
            "Min": 180,
            "Max": 190 
        }
    ],

    "treshold": 80
}

constraints <- jsonlite::read_json("constraints.json",simplifyVector=TRUE)

Cal file:

Martonvasar_maize.obs
Martonvasar_maize.set
site
Martonvasar_maize;211