Skip to content

Latest commit

 

History

History
75 lines (51 loc) · 4.23 KB

README.md

File metadata and controls

75 lines (51 loc) · 4.23 KB

WRF-SUEWS Pre-processor (WSPS)

Steps in summary

  1. Use conda to create a fresh environment use environment.yml:
conda env create -f environment.yml

  1. Set up WSPS configuration in the wsps section of namelist.suews, and add SUEWS related run files associated with the configuration in the spin up folder. You also need to add all wrfinput_d0*.nc files from WPS process to your input folder

  2. run:

python wsps.py

  1. The modified wrfinput_d0*.nc files will be in the generated output folder output/final.

  2. (optional) If you wish to have site-specific modification of wrfinput.nc files (e.g. land cover, population density etc.), you can use wps_site_specific.py script in conjunction with customised related modules in utility folder (see examples of London and Swindon):

python wsps_site_specific.py

Steps in detail

Mandatory steps (general)

Configuration

In the first step, the WSPS needs to be configured in wsps section of namelist.suews (in the root directory of pre processor folder). Please read below to see how to modify configuration (and add necessary SUEWS related files) for your purpose:

&wsps
urban_site_spin_up    = 'London', 'Swindon'
urban_domain_number   = '03',     '04'
urban_class_threshold = 0.6,      0.16
urban_class           = 'London', 'Swindon'
veg_site_spin_up      = 'Swindon'
transmissivity        = 0.1257,   0.1216
start_date            = '2012-01-10'
output_file_name      = 'output'
input_file_name       = 'input'
SUEWS_param_template  = 'SUEWS_param.json'
phenology_parameters  = 'phenol_attrs.csv'
data_dir              = 'data'
/

Here are the explanation of each option:

  • urban_site_spin_up: this is the list of the urban site you like to use for spin up. For each of the name in this list, there should be a same-named folder with SUEWS related files to run SuPy in the spin up folder.

  • urban_domain_number: this is the domain associated with each of sites in urban_site_spin_up

  • urban_class_threshold and urban_class : these are threshold related to urban classes (urban fraction) and the associated site for each one. You can have as many threshold as you want. For example, in the above script, the thresholds are 1 < f < 0.6 is assigned to London, 0.6 < f < 0.16 is assigned to Swindon, and the rest are automatically assigned to vegetated classes.

  • veg_site_spin_up: this is the site configuration you need to use for vegetated spin up. It can be any configuration as long as there is a same-named folder with SUEWS related files to run SuPy in the spin up folder. In the example above, Swindon site configuration is used.

  • transmissivity: The values for transmissivity correction for each urban site. Put them 0 if no correction is needed.

  • start_date: Start date of the run

  • output_file_name: the output file name that will be generated in sample case folder

  • input_file_name: the output file name contains necessary inputs in sample case folder

  • SUEWS_param_template: the name of json file containing all the SUEWS variables needed to be added to WRF inputs (it is in the input folder).

  • phenology_parameters: The name of the file containing parameters related to non-urban areas such as albedo, LAI, conductances based on Omidvar et al. 2020 (it is in the input folder)

  • data_dir: name of the data directory necessary for site_specific modifications (it is in the sample case folder)

Site specific steps

If you wish to have site-specific modification of wrfinput.nc files (e.g. land cover, population density etc.), you can use wps_site_specific.py script in conjunction with customised related modules in utility folder. You need to follow this template for writing the script for modification of your own site. see examples of London and Swindon in the utility folder.