Weather and Climate Prediction Exercise 5
Due Friday Feb 11

In this exercise you will learn about how fortunate we are to have a well behaved planet.


I. Build and Run CAM (~5 min, plus a long wait time)

II. Meanwhile analyze CAM output from a couple of runs Cecilia has done aready in MATLAB. By Monday you can analyze runs done by you and your classmates. Also continue monitoring Cecilia's somplanet run to see if it goes into a snowball earth.

I. Build and Run CAM.

Rob and Josh will run a case with the earth tilted at 60 deg rather than 23 deg. Kris and Alex will run a case with the earth in a synchronous orbit around the sun (the rotation rate is 1/365 of normal and the sun always lights the same side of the planet). If Bill is here, he will run an earth tilted at 45 deg. Your runs will take about a day to complete. They may blow up, which of course won't be counted against you. We will just analyze them to the point where they blow.

Rob and Josh - do this!

cd /home/disk/p/atms380/$LOGNAME/camruns
mkdir tilted60
cd /home/disk/p/atms380/$LOGNAME/camruns/tilted60
cp /home/disk/p/atms380/scripts/bld-tilted60.csh .

Kris and Alex - do this! (oops asynch should be synch)

cd /home/disk/p/atms380/$LOGNAME/camruns
mkdir asynchlowf
cd /home/disk/p/atms380/$LOGNAME/camruns/asynchlowf
cp /home/disk/p/atms380/scripts/bld-asynchlowf.csh .

Bill, if you are here, do as Rob and Josh, but substitute 45 for 60

All of these runs are done with a Slab Ocean Model (SOM), which has a fixed 50 m depth. There is no land and no topography. I have already started a zero obliquity run that I am calling somplanet.

Now everyone: Wait a few minutes. When done submit the job.

qsub run-cam3.csh
   Verify your job is in the queue with either
qstat -u "*"
the latter lets you see all the jobs in the queue.

   If for some reason you wish to kill your job.
qdel xxxxx                     fill in the x's with the job-ID and this will cancel your job

II. Analyze CAM

  Make a supdirectory for this exercise's analysis files. Maybe you would like to call it "weirdplanets". Go to that directory and copy the analysis files to your directory for this exercise. Start matlab

cd /home/disk/p/atms380/$LOGNAME/camruns
mkdir weirdplanets
cp /home/disk/p/atms380/scripts/analyze_ex5* .
cp /home/disk/p/atms380/scripts/globmean.ncl .
matlab &

  a) Run analyze_ex5_a in matlab and choose Modern. Ah! The home planet - with continents, mountains, and the right obliquity. Look at your favorite variables. For sure, check out the surface temperature (T), sea ice fraction (I), and vertical profile of the zonal wind (U).

Beware! Not all the planet runs are done! Yours for instance is just starting to run. Choosing an unfinished run will probably make a figure or two and then a lot of error messages. The error messages are harmless. Just keep restarting the script and try other options

  To turn in: Compare and constrast the vertical structure of the jet via the zonal wind (U) with the vertical structure of temperature (T) for each run (modern, somplanet, tilted60, and asynchlowf; tilted90 is optional).

  b) Next compute the global mean temperature and northern hemisphere sea ice area.First we will do it for the somplanet run. Later you can edit the script and make it read your run. Run these commands IN A TERMINAL NOT IN MATLAB:

cd /home/disk/p/atms380/$LOGNAME/camruns/weirdplanets

setenv NCARG_ROOT /usr/local/ncarg-5.1.0
/home/disk/sys/local/linux64/ncarg-5.1.0/bin/ncl < globmean.ncl

This will create a couple files like Tglob_somplanet.txt, icenh_somplanet.txt
Read them into matlab by hand with "load Tglob_somplanet.txt". Plot in matlab with

To turn in: Describe global mean temperature and hemspheric sea ice area for the somplanet run. Do the same for tilted60 and asynchlowf. Do any of the runs exhibit instablities? Run analyze_ex5_a.m and illustrate the character of instabilities if they occur. If they do not, choose something else, like the general seasonal cycle and extremes to discuss.

Return to Homepage