WGEN

Weather generation scripts based off of the Richardson (1984) climate model WGEN. Generates daily weather data for rainfall, minimum temperature, maximum temperature, and solar radiation from input daily weather data.

How to Use

Inputs:

• Input weather dataset (csv): daily values - sample input data set provided in repository
  • Necessary variables:
    • "month" month (1-12)
    • "year" year
    • "day" day (1-28,30,31)
    • "pred" rainfall values (mm)
    • "tmnd" minimum temperature (°C)
    • "tmxd" maximum temperature (°C)
    • "solar" solar radiation (MJ/m**2/d)
• Monthly means and sds (xls): monthly averages of each weather variable that the generated dataset should target - sample input file provided in repository
  • "month" month (1-12)
  • "rainmn" mean daily rainfall
  • "rainsd" standard deviation of daily rainfall
  • "tminmn" mean minimum temperature
  • "tminsd" standard deviation of minimum temperature
  • "tmaxmn" mean minimum temperature
  • "tmaxsd" standard deviation of minimum temperature
  • "sradmn" mean solar radiation
  • "sradsd" standard deviation of solar radiation
  • "tnmnd" mean minimum temperature for only dry days
  • "tnsdd" standard deviation of minimum temperature for only dry days
  • "tnmnw" mean minimum temperature for only wet days
  • "tnsdw" standard deviation of minimum temperature for only wet days
  • "txmnd" mean maximum temperature for only dry days
  • "txsdd" standard deviation of maximum temperature for only dry days
  • "txmnw" mean maximum temperature for only wet days
  • "txsdw" standard deviation of maximum temperature for only wet days
  • "xmnd" mean solar radiation for only dry days
  • "xsdd" standard deviation of solar radiation for only dry days
  • "xmnw" mean solar radiation for only wet days
  • "xsdw" standard deviation of solar radiation for only wet days
  • "rnum" mean number of wet days

Quick Start

1. Set repository WGEN as directory
2. Place the input historical dataset labelled and formatted as specified above in the "Inputs" folder of the repository
3. Generate the "Input weather weather dataset" input file using the "compile_means_sds.py" script
4. Open "wgen.py"
5. In section 1, fill in the variables listed
   1. input_mns_sds = name of the excel file that holds the input values for means and sds (make sure to type in with quotation marks and add file extension .xls 'string')
   2. input_data = name of the csv file with input dataset (type in with quotation marks and add file extension .csv 'string')
   3. strt_yr = start year
   4. n_yrs = number of years you want to generate
   5. corrections = True or False (It is recommended to apply corrections to generated dataset for more accurate results)
   6. output_name = name output csv file (type in with quotation marks and add file extension .csv 'string')
6. Run script
7. Resulting output data will be located in the "Outputs" folder of the repository

Definitions

• A day is considered wet if rainfall for the day is at least 0.01 mm
• Please be mindful that as of now, the generation will always start at January!