UBC ATSC 507 - Numerical Weather Prediction (NWP)

Running WRF

Under construction (this web page, and most other web pages for this course).


Instructor    Tim Chui

Learning Goals:    By the end of this module, you will be able to ...

Readings BEFORE class:

  1. See Lecture Notes from ATSC 212 on Linux:
    https://www.eoas.ubc.ca/courses/atsc212/labs/linux-1/UNIXprint.pdf .
  2. Although not assigned here, if you want, you can practice your Linux skills with the homework at
      https://www.eoas.ubc.ca/courses/atsc212/labs/linux-1/ 

IN class: Bring your laptops (with chargers if needed), to use as a remote terminal to access the computer cluster named "Optimum".

AFTER class, Homework #2 , due in roughly 1 week:

  1. Create a WRF simulation of the 2018-12-20 wind storm case study using the "benchmark" PBL scheme that Tim demonstrated in class.
  2. Create another WRF simulation of the same 2018-12-20 wind storm case study, but this time using a different PBL scheme, assigned as below.
    Here are some references for physics schemes and Planetary Boundary Layer (PBL) options in WRF:
    • References for physics schemes: https://www2.mmm.ucar.edu/wrf/users/physics/phys_references.html#PBL
    • PBL Options, WRF User's Guide Chapter 5: https://www2.mmm.ucar.edu/wrf/users/docs/user_guide_v4/v4.4/users_guide_chap5.html#summary

  3. Using IDV or Panoply (or whatever visualization software you prefer), create several plots of standard variables (like temperature, precipitation, winds, MSLP, etc.). Compare the plots of your specific simulation to that of the benchmark simulation that we did in class. Ensure that the ranges you set are identical when comparing plots (i.e. explicitly set ranges, do not autorange).
  4. In a few sentences, describe what PBL parameterizations do within the WRF model. In another few sentences, and using the journal papers associated with your assigned PBL scheme, provide a short description of your scheme. In particular, discuss the closure of the scheme, and how it handles turbulent quantities like Turbulence Kinetic Energy. In total, this should take up a single paragraph. No equations need to be involved.
  5. Write up a short, informal, and entirely qualitative report discussing differences you see by eye between the simulations, supported by your plots. For example, when does it become apparent that the precipitation fields differ between your simulation and the benchmark simulation, if at all? If there are differences in the locations of certain precipitation cells, where do they occur in your domain?  What factors might cause these differences to occur? In total, this should take up a handful of pages at most, including figures and discussion.


Topics

A.  Review of Linux, bash, etc.

  1. See Lecture Notes from ATSC 212 on Linux:
    https://www.eoas.ubc.ca/courses/atsc212/labs/linux-1/UNIXprint.pdf .
  2. Here is a fun little demo on how namelists work in Fortran. namelist_demo.zip , includes the fortran code, input files, and sample output.

B.  WRF - Part 1: Install

  1. See Tim Chui's WRF tutorial #1 on how to install WRF on UBC's "Optimum" computer.

C.  WRF - Part 2: Run

  1. See Tim Chui's WRF tutorial #2 on how to run WRF on Optimum.

D.  WRF - Part 3: Visualization of Outputs

  1. IDV:  See Tim Chui's WRF tutorial #3 on how to use IDV to display your output.
  2. Panoply (from NASA): See Tim Chui's panoply_wrf.pdf tutorial.
  3. Vapor (from NCAR) , with installation instructions and tutorial by Dr. Nadya Moisseeva, with a large  sample NetCDF file   or a smaller .vdc file you can download to view in vapor.
  4. Others?