Monthly Archives: September 2011

i found a hint http://www.atmos.washington.edu/academics/classes/2011Q1/380/MessWithX.html yes this might actually solve all my issues

i’ve compiled CAM4 before – I’ll just try it again and see what happens. (with ALL the previous combinations i’ve tried out, too)

i tried changing dynamics to both semi-lagrangian and finite-volume unfortunately, both failed almost immediately if i do that, i’d have to change a bunch of other things around too maybe changing the grid size would work better.

dtime The length (in seconds) of the coupling interval between the dynamics and physics. Changing this variable directly impacts the physical parameterizations in the model and may impact the climate. Changing resolution usually requires a change in dtime. Default: is … Continue reading →

what i also want: a mouseover button for text editors (that auto-converts anything math-ish to LaTeX format) == good design: good variables are ctrl-F-able across multiple files (has anyone ever thought of this?) == it may not be wise to … Continue reading →

http://kingkong.amath.washington.edu/uwamath583/sphinx/notes/html/fortran_sub.html#subroutines man i forget everything don’t i?

this will be pretty fun though

call ESMF_TimeGet(date, yy=year, mm=month, dd=day, s=tod, rc=rc) ^i have to change year and day before this is called, I think == i still need to understand private/public scope in fortran better

if heat capacity and dissipative coefficients are given in terms of years, then dividing them by 86400 actually shouldn’t do any harm at all since we want them to be in terms of seconds anyways. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.76.84&rep=rep1&type=pdf i can’t be TOTALLY … Continue reading →

so i really read over the CAM3 manual this time (not very thoroughly, but still much more so than before) and finally have an idea of what these equations are doing too bad the values of 86400 aren’t glued to … Continue reading →

so… i actually had the idea: what if i changed BOTH the coriolis effect 86400 *and* the zenith angle one? except – there are still issues with 365 being hard-coded in everywhere but the scaling factor the NASA scientist told … Continue reading →

the surface radiation mod too

qflux.c also has a value of 86400 in it. but cecilia’s suggestion was to set qflux to 0 so maybe that won’t be a condition for CAM3. For CAM5 – i’ll have to look further

dissipation constants include 86400 for both wind and temperature dissipation. i’ll have to read the below paper to see if there’s anything physically significant about that 86400 ! Held, I. M., and M. J. Suarez, 1994: A proposal for the … Continue reading →

perpetual logical Set to TRUE to specify that the run will use a perpetual calendar, i.e., a diurnal cycle will be present for the fixed calendar day specified by perpetual_ymd. perpetual_ymd integer Perpetual date encoded in an integer as (year*1000 … Continue reading →

i should NOT obsess over the value of 86400 in each and every file. i just need to change it first in the time_manager mod first. then maybe everything will work if i masschange that 86400 value everywhere else

I seriously did NOT remember how fortran returns its values until now. no, the language DOES specify which variables it returns in its header. fortran DOES allow you to change the arguments in here after all

i took a short break to fill in the gaps I had from C, but now i’m back again, looking over the CAM5 code. ESMF_TimeMod.F90 seems to be where a lot of the time calculations are done. ESMF_TimeMgr_INC is where … Continue reading →

I’m actually just reading K&R’s C Programming Language today – it should get me to read the C code in the netcdf C functions (which I had a bit of trouble reading due to my inexperience with pointers and structures)

NASA scientist told me to use CAM5 instead. so i decided to take the plunge (for now). its time_manager.F90 doesn’t have 86400 scattered throughout it. which is good. but it still has lots of hard-coded things scattered throughout the code. … Continue reading →

http://books.google.com/books?id=dw2GadaPkYcC&pg=PA556&lpg=PA556&dq=obliquity+of+the+early+earth&source=bl&ots=TrF8IS073k&sig=-cf4NSdImJv12XOUEeAvtxQinH0&hl=en&ei=oNZvTqrwC6fUiALLrsH2Bg&sa=X&oi=book_result&ct=result&resnum=3&ved=0CC8Q6AEwAg#v=onepage&q&f=false http://www.nature.com/nature/journal/v396/n6710/abs/396453a0.html == so… 23.5 degrees is actually quite arbitrary. maybe it would be simpler to do it for a range of obliquities, or just one of 0. still, 23.5 is easiest to interpret, and would help prevent certain undesirable … Continue reading →

http://faculty.plattsburgh.edu/mary.rodentice/courses/petrolpresent/Ch%2017%20Cont%20Arcs%20’07.pdf okay, emailed some more professors. so now it appears that it’s far more than just island arcs – as the Archean land distributions are not particularly well-constrained. == i did just become a reviewer on Quora, and got a … Continue reading →

i think i’ll just read global physical climatology today. it’ll just take a day to get through most of the book anyways.

changing the day length somehow screws up the time_const.f90 to the point that one of the variabels in time_const isnt even called, and can’t be passed to that silt routne. aanyways i’ll take a look when i wak eup

the code has now SUDDENLY become far more manageable this way!

when i was looking through the code, i didn’t even TRY to see whether i was looking at lagrangian or eulerian dynamics. well – turns out – that i should have looked at the stuff in eulerian dynamics i don’t … Continue reading →

ifort -c -FR -I. -I/home/disk/p/atms380/achen89/camruns/rotate2xReplaceTimeManager/SourceMods -I/home/disk/eos11/bitz/cam3.1/cam1/models/atm/cam/src/physics/cam1 -I/home/disk/eos11/bitz/cam3.1/cam1/models/atm/cam/src/dynamics/eul -I/home/disk/eos11/bitz/cam3.1/cam1/models/atm/cam/src/control -I/home/disk/eos11/bitz/cam3.1/cam1/models/csm_share/shr -I/home/disk/eos11/bitz/cam3.1/cam1/models/atm/cam/src/utils -I/home/disk/eos11/bitz/cam3.1/cam1/models/utils/timing -I/home/disk/eos11/bitz/cam3.1/cam1/models/atm/cam/src/advection/slt -I/home/disk/eos11/bitz/cam3.1/cam1/models/atm/cam/src/ocnsice/som -I/home/disk/eos11/bitz/cam3.1/cam1/models/lnd/clm2/src/main -I/home/disk/eos11/bitz/cam3.1/cam1/models/lnd/clm2/src/biogeophys -I/home/disk/eos11/bitz/cam3.1/cam1/models/lnd/clm2/src/biogeochem -I/home/disk/eos11/bitz/cam3.1/cam1/models/lnd/clm2/src/mksrfdata -I/home/disk/eos11/bitz/cam3.1/cam1/models/lnd/clm2/src/riverroute -I/home/disk/eos11/bitz/cam3.1/cam1/models/ice/csim4 -I/usr/include -I/usr/rels/open-mpi/include -I/home/disk/p/atms380/achen89/camruns/rotate2xReplaceTimeManager/bld/esmf/mod/modO/linux_intel -I/usr/include -DCAM -DNO_SHR_VMATH -DHIDE_SHR_MSG -DNO_R16 -DLINUX -132 -autodouble -O2 /home/disk/eos11/bitz/cam3.1/cam1/models/atm/cam/src/dynamics/eul/scanslt.F90 /home/disk/eos11/bitz/cam3.1/cam1/models/atm/cam/src/dynamics/eul/scanslt.F90(149): error #7002: Error in opening the compiled … Continue reading →

nelapse: Elapsed time to run. In time steps (positive) or days (negative) to run. May be entered instead of NESTEP orSTOP_YMD and STOP_TOD. Note, NESTEPis not required or used when running as part of CCSM. Setting is Required if NESTEP … Continue reading →

went into the configure list, found that our model uses eulerian dynamics. so 4.1.2 dtime Type integer Default 1200 s (Eulerian dycore), 1800 s (Finite-volume dycore), 3600 s (SLD dycore) Use Timestep size in seconds. dtime must evenly divide the … Continue reading →

permissions have nothing to do with it

now i’ve just realized that it might actually be the permissions. unlikely, but i’ll try anyways

TIL that the name of the bld csh file doesn’t mean anything at all. (yeah that was late)

just tried it without adding all the bld files. mean molar constant still fails (with exact same errors as before)

i think i know what the problem is. it’s with make. and the thing with make is this: i’m was COPYING the tilt60 folder EVERYWHERE just to create new simulations. and that INCLUDED files built by the old tilted60 planet. … Continue reading →

atms380 student (reb42) ONLY included shr_const_mod.F90 in sourcemods directory. i included a bunch of other stuff. i should try removing that

4.1.2 dtime Type integer Default 1200 s (Eulerian dycore), 1800 s (Finite-volume dycore), 3600 s (SLD dycore) Use Timestep size in seconds. dtime must evenly divide the number of seconds in a day.

********** Time Manager Configuration ********** Calendar type: NO_LEAP Timestep size (seconds): 1800 Start date (ymd tod): 5030101 0 Stop date (ymd tod): 5130101 0 Reference date (ymd tod): 5030101 0 Current step number: 0 Ending step number: 175200 Current date … Continue reading →

change in mean molar constant: here’s the end: COURLIM: *** Courant limit exceeded at k,lat= 1 25 (estimate = 1.014), solution has been truncated to wavenumber 30 *** *** Original Courant limit exceeded at k,lat= 1 25 (estimate = 1.014) … Continue reading →

did not realize that there was an entire document devoted to the time-manager module http://www.cesm.ucar.edu/models/atm-cam/docs/time-manager/time-manager.html i should read that == also i’m now analyzing cam.out (with the mean molar mass change) to see what went wrong. the cam.out file is … Continue reading →

time_manager.o : $(ESMF_LIB)/$(ESMF_ARCH)/libesmf.a

nstep = get_nstep() == it crashes when it calls that from the time_manager.f90 file. now i need to check make dependencies and see if what else depends on that. and the first f90 file that depends on time_manager

make: Circular scanslt.o <- time_manager.o dependency dropped. == i’ll have to google scanslt

just used Notepad++ to compare the make.out (with gas constant change) and make.out (with day length change). absolutely no difference

Changing 86,400 in shr_const_mod.F90 works. But here’s the thing: changing the gas constant in the same file – does not cause the build to fail – but then run-cam3.csh fails to produce any output at all. gah Anyways – I’m … Continue reading →

Figure 3: Extent of upstream cold region as a function of planetary rotation rate. The extent of the upstream cold region is the longitudinal distance between the centre of the triangle and the westernmost extent of the bold −2 K contour … Continue reading →

In middle latitudes, the eastern boundaries of Northern Hemisphere continents are colder than other regions at similar latitudes. Northeastern North America in winter is up to 20 K colder than similar latitudes in Europe, and northeastern Asia is similarly cold in … Continue reading →

Winter cold of eastern continental boundaries induced by warm ocean waters Yohai Kaspi & Tapio Schneider Affiliations Contributions Corresponding author Nature 471, 621–624 (31 March 2011) doi:10.1038/nature09924 Received 30 August 2010 Accepted 10 February 2011 Published online 30 March 2011 … Continue reading →

This quote taken explicitly from the Venus GCM paper: "The decadal periodicity exhibited by the model may or may not represent a realistic stte of the Venus atmosphere since we have made a number of simplifying assumptions in our simulations, … Continue reading →

http://www.nature.com/news/2011/110902/full/477145a.html

i’m going to the university tomorrow, and I am *definitely* going to print out the Venus GCM pdf, as well as the netCDF tutorial PDF. sometimes I wish I had a laser printer at home

i didn’t post any answers on Quora today, which is good, but I did get a bit carried away by http://www.quora.com/Edwin-Chen-1/Dissecting-the-Spread-of-a-Quora-Post-Information-Transmission-in-a-Social-Network . This is also very exciting potential research, but I shouldn’t get too carried away. Anyways, I actually just … Continue reading →

if i am to simulate the early earth, i’m also going to have to dial down the solar constant but here’s the thing: it also means that i have to crank up CO2 levels somewhere. hopefully without having to hack … Continue reading →

i might not make some progress for a while, because i need to learn how to work with netcdf scripts so now i’m really perusing the NetCDF tutorials http://www.unidata.ucar.edu/software/netcdf/docs/netcdf-tutorial.html#Common-Data-Model http://www.unidata.ucar.edu/software/netcdf/docs/netcdf-tutorial.html#Data-Model http://www.unidata.ucar.edu/software/netcdf/examples/files.html

got frustrated with network connectivity issues, so i finally took the matter in my own hands and learned as much as i can about network speeds. anyways, http://www.quora.com/How-does-the-distance-to-a-particular-city-affect-the-ping-and-bandwidth-of-your-Internet-connection-to-that-city shows the graph i ultimately managed to produce. on the other hand, … Continue reading →

so i now know that time_manager.f90 really was the culprit. So this time, I might as well try copying ALL the files (including unmodified ones) over. then I may be able to recompile and get it to work.

so this time i decided to ONLY change both the constants file and time-manager.F90. unfortunately, it turns out that time-manager.F90 might be the culprit of the crash. unfortunately, it was the most critical part of it. when i was going … Continue reading →

yes this is where grid size matters

it’s NOT just the Coriolis effect of rotation. it’s also the solar zenith angle. uh oh, i’m scared that the grad student may have forgotten that in his publication zenith has jdays as an argument. jdays relies on get_curr_calday, which … Continue reading →

i wonder waht that means

zenith angle is used for shortwave radiation calculations, meaning that it’s super-important so next step is to (again) try to find if there’s any dependency between zenith angle and shr_const_mod’s value for number of seconds in a day

!

so the error screwed plans, and i’ll try to analyze what went wrong in the error output file tomorrow. for now, i’ll just run 3 simulations overnight (a) change 86400 (b) change mean molar mass to 44 (CO2 value) (c) … Continue reading →

32 icicle% bld-rotate2xReplace86400.csh /home/disk/p/atms380/akchen0/camruns/rotate2xReplace86400 creating /home/disk/p/atms380/akchen0/camruns/rotate2xReplace86400/bld/Filepath creating /home/disk/p/atms380/akchen0/camruns/rotate2xReplace86400/bld/params.h creating /home/disk/p/atms380/akchen0/camruns/rotate2xReplace86400/bld/misc.h creating /home/disk/p/atms380/akchen0/camruns/rotate2xReplace86400/bld/preproc.h creating /home/disk/p/atms380/akchen0/camruns/rotate2xReplace86400/bld/Makefile creating /home/disk/p/atms380/akchen0/camruns/rotate2xReplace86400/bld/config_ca che.xml configure done. building CAM in /home/disk/p/atms380/akchen0/camruns/rotate2xReplace86400/bld .. . CAM build failed: see /home/disk/p/atms380/akchen0/camruns/rotate2xReplace86400/ bld/MAKE.out

I have PowerGREP to do it on Windows. But I need to learn how to do it on UNIX too. Anyways… http://kb.iu.edu/data/afrk.html If you have Perl installed on your system With Perl, you can make the replacement from the Unix … Continue reading →

"Effects of Initial Flow on Close-In Planet Atmospheric Circulation" http://arxiv.org/PS_cache/arxiv/pdf/1004/1004.2871v1.pdf I don’t think he changed things as comprehensively as I did. But I’ll see what I can get.

So.. A problem is that the code will not be self-consistent in radiation, e.g. #1, an ozone layer might be hard-wired into radiative calculations, which doesn’t apply to an anoxic Archean Earth, e.g. #2, greenhouse warming assumptions for the modern … Continue reading →

so while i’m still trying to understand the code and the dependencies, there are at least a few "brute force" things I can do that aren’t really brute force I’m going to try to run several different simulations of this … Continue reading →

when i’m doing mass searching, i have to note what folder the constant is in. If it’s in the physics or dynamics folder, then it’s probably important. if it’s in some other folder, maybe not so much (yet)

On top of that evil number 86400… there are lots of places in the code where you have to deal with "365", 365 being the number of days in a year there’s one place in the code where you have … Continue reading →

haha. I didn’t get too much done today (Quora addiction and some issues with low energy levels), but I’m still getting some good ideas. === So… Regarding the generation of landmasses relevant for CAM, I actually figured that some pretty … Continue reading →

The Hunt for Other Earths Begins at Home Planet Earth is the only example we have of what a habitable planet ‘looks’ like. Using observations from NASA’s Terra, QuikSCAT, and Aura missions, researchers have now developed a 3-D Spectral Earth … Continue reading →

Ian said that “island arcs” are what we’re most interested in. But that’s the issue here – island arcs, by nature, are difficult to generate through some non-arbitrary process. You can manually create them, but then how do you know … Continue reading →

also, the last post very well applies to all the other constants that were scattered throughout the program too. if it does, then maybe the complication i faced might not be that bad at all.

i need to sleep == i think i know what the issue now is. the shr_const_cday variable might not even be *used* by the zenith function. the zenith function apparently uses an argument of 86400 defined somewhere else in the … Continue reading →

http://stackoverflow.com/questions/7296752/are-there-any-fortran-compilers-viewers-that-let-you-see-the-dependencies-of-a-gi And what does it look like? This: So that way, I can try tracking dependencies. For much of today, I tried to find where the zenith function interacted with shr_const_mod.F90. So far, however, I have been unsuccessful. But I … Continue reading →

So… after a helpful email from some U.K researchers, I managed to get this… And so I got excited and thought – hey – maybe my problems are now all solved! Except… that many of the constants are actually hard-coded … Continue reading →

Welcome to WordPress.com. After you read this, you should delete and write your own post, with a new title above. Or hit Add New on the left (of the admin dashboard) to start a fresh post. Here are some suggestions … Continue reading →