From kfischer@cgd.ucar.edu Fri Jan 10 09:59:58 2003 Received: from olympic.cgd.ucar.edu (olympic.cgd.ucar.edu [128.117.23.81]) by bearmtn-e0.cgd.ucar.edu (8.9.3/8.9.3) with SMTP id JAA27637 for ; Fri, 10 Jan 2003 09:59:58 -0700 (MST) Message-Id: <200301101659.JAA27637@bearmtn-e0.cgd.ucar.edu> Date: Fri, 10 Jan 2003 09:58:12 -0700 (MST) From: Kathy Fischer Reply-To: Kathy Fischer To: ccsm-crb@bearmtn-e0.cgd.ucar.edu MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: EAcaV0MLroCeQzrFBnrI/w== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.3.4p_5 SunOS 5.7 sun4u sparc Subject: [CCSM-CRB] test Sender: ccsm-crb-admin@goldhill.cgd.ucar.edu Errors-To: ccsm-crb-admin@goldhill.cgd.ucar.edu X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.0.1 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: testing ... Sorry for the interuption. Kathy =~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~= Kathleen Fischer (303) 497-1644 NCAR - CGD Information Systems kfischer@ucar.edu =~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~= No eternal reward will forgive us now for wasting the dawn. -Jim Morrison From lydia@ucar.edu Thu Feb 13 17:54:49 2003 Received: from bearmtn-e0.cgd.ucar.edu (bearmtn [128.117.24.2]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h1E0snul004361 for ; Thu, 13 Feb 2003 17:54:49 -0700 Received: from ucar.edu (ras71.ucar.edu [128.117.68.71]) by bearmtn-e0.cgd.ucar.edu (8.9.3/8.9.3) with ESMTP id RAA23576; Thu, 13 Feb 2003 17:57:30 -0700 (MST) Message-ID: <3E4C3E36.926C1497@ucar.edu> Date: Thu, 13 Feb 2003 17:54:14 -0700 From: Lydia Harper Organization: Climate and Global Dynamics Division X-Mailer: Mozilla 4.06 [en] (WinNT; I) MIME-Version: 1.0 To: CCSM-CRB@goldhill.cgd.ucar.edu Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit cc: Lydia Harper Subject: [CCSM-CRB] Notes from CCSM-CRB meetings X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 14 Feb 2003 00:54:49 -0000 Hi, Here is the URL for the CCSM-CRB home page: http://www.ccsm.ucar.edu/management/CCRB/ Notes from the first and second meetings can be viewed here. Thanks, Lydia From tcraig@ucar.edu Fri Feb 28 12:08:18 2003 Received: from bearmtn-e0.cgd.ucar.edu (bearmtn [128.117.24.2]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h1SJ8Iul021596 for ; Fri, 28 Feb 2003 12:08:18 -0700 Received: from sunray2.cgd.ucar.edu (sunray2 [128.117.24.201]) by bearmtn-e0.cgd.ucar.edu (8.9.3/8.9.3) with ESMTP id MAA00571 for ; Fri, 28 Feb 2003 12:08:15 -0700 (MST) Received: from localhost (tcraig@localhost) by sunray2.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h1SJ8Fq15527 for ; Fri, 28 Feb 2003 12:08:15 -0700 (MST) X-Authentication-Warning: sunray2.cgd.ucar.edu: tcraig owned process doing -bs Date: Fri, 28 Feb 2003 12:08:15 -0700 (MST) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] status of cpl6 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Feb 2003 19:08:19 -0000 CCSM-CRB, The ccsm-crb meeting scheduled for this week was preempted by the scientist meeting. As a result, we were not able to talk about the status of cpl6. Maybe we can handle this via email. Results from cpl6 were presented this week at both the scientists' and programmers' meetings. Both Brian and I feel (and hopefully showed) that cpl6 is duplicating cpl5 science to round-off except in places where cpl5 bugs were fixed. We feel that the differences in cpl6 compared to cpl5 are orders of magnitude smaller than the scientifically insignificant bugs we found in cpl5 as a result of this process. In other words, we think cpl6 is ready for production. Our final validation test is a continuation of the b20.007 control run for a few decades. b21.002 was started this week from date 0950-01-01 of b20.007 (the "1000" CCSM2.0 control). Early next week, we plan to start looking at the results. It has already run about a decade, and we don't anticipate any problems. We are in the process of validating cpl6 on the SGI and we will be adding some diagnostic features to cpl6 in the next week. These outstanding requirements should be resolved soon and have no impact on science or on the ability to run CCSM2.1 on the IBM. I would like to propose that the CCSM Software Engr Group start using cpl6 versions of CCSM components for all future development, starting next week, if nothing scary shows up in the b21.002 test. In other words, starting next week, I'd like our default CCSM development tags to include cpl6 and that versions including cpl5 will not longer be on the CCSM development path (although cpl5 will continue to be supported as part of past releases). If you have any suggestions about other tests that should be done, or concerns that cpl6 isn't ready, please respond to the mailing list, ccsm-crb@cgd.ucar.edu or to me directly. Thanks, tony............ From tcraig@ucar.edu Tue Mar 11 14:02:52 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h2BL2qR2014245 for ; Tue, 11 Mar 2003 14:02:52 -0700 Received: from sunray3.cgd.ucar.edu (sunray3.cgd.ucar.edu [128.117.24.203]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h2BL2HKm005080 for ; Tue, 11 Mar 2003 14:02:17 -0700 (MST) Received: from localhost (tcraig@localhost) by sunray3.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h2BL2HE07443 for ; Tue, 11 Mar 2003 14:02:17 -0700 (MST) X-Authentication-Warning: sunray3.cgd.ucar.edu: tcraig owned process doing -bs Date: Tue, 11 Mar 2003 14:02:17 -0700 (MST) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] Meeting Tomorrow, Wed 3/12 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 11 Mar 2003 21:02:52 -0000 CCSM-CRB meeting Wednesday, March 12 2pm Room CG1 3416 I'd like to propose a CCSM-CRB meeting tomorrow at 2pm in CG1 3416. I think there are a number of things to review - Status of cpl6 - Incorporation of rain/snow coupling modifications - Incorporation of Bill Large coupler/coupling modifications - Other IPCC modifications - IPCC plans - What is expected on April 8? - How will modifications in different components be integrated into CCSM? Will there be a formal process? Do we need to schedule? Is there a test plan? Sorry for the late notice. Thanks, tony............ From tcraig@ucar.edu Thu Mar 13 16:23:23 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h2DNNNR2015043 for ; Thu, 13 Mar 2003 16:23:23 -0700 Received: from sunray3.cgd.ucar.edu (sunray3.cgd.ucar.edu [128.117.24.203]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h2DNMhKm011898 for ; Thu, 13 Mar 2003 16:22:43 -0700 (MST) Received: from localhost (tcraig@localhost) by sunray3.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h2DNMhG03308 for ; Thu, 13 Mar 2003 16:22:43 -0700 (MST) X-Authentication-Warning: sunray3.cgd.ucar.edu: tcraig owned process doing -bs Date: Thu, 13 Mar 2003 16:22:43 -0700 (MST) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] cpl6 runoff questions X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 13 Mar 2003 23:23:23 -0000 CCSM-CRB, A bit more on the runoff differences between b20.007 and b21.002. This was brought up yesterday at the CRB meeting with respect to the long cpl6 validation. I have verified that for the 10 year average period 960-969, b20.007 has 1.388e9 kg/s, global sum of water runoff in ocean b20.002 has 1.384e9 kg/s, global sum of water runoff in ocean so the total amount of runoff going into the ocean is very nearly the same. The coupler is conserving runoff between land and ocean on a timestep by timestep basis based on the coupler diagnostics. The regional differences in ocean runoff are typically no more than 10% to 20% in regions with significant runoff. The regional differences in runoff are consistent between the land history and ocean history files suggesting no problem in the coupling of the runoff field. The regional differences in runoff are consistent with changes in the precip pattern as seen in the land history files. My current feeling is that the coupler is handling runoff just fine. The differences in runoff seen in the ocean are consistent with the changes in the precip differences in these 10 year averages. The precip differences are probably the result of variability in the model on decadal timescales, possibly slight differences in "physics" (all small bug fixes) between the b20.007 control and the latest b21.002 case, or possibly differences in the initial condition of the runs. b21.002 did not start from a spun up land state. I am setting up 2 new cases, b20.022 and b21.004 - They will be basically the same components except for the coupling interface. these runs are based on ccsm2_0_beta65 and ccsm2_1_beta12. b20.022 b21.004 atm: cam2_0_1_dev15 cam2_0_1_dev15_brnchT_cpl6_a030214 lnd: clm2_deva_08 clm2_deva_11_brnchT_cpl6_2 ice: csim4_7_0 csim4_7_0_brnch_cpl6_4 ocn: ccsm_pop_1_4_20030117 ccsm_pop_1_4_20030117_br_cpl6_20030130 - They will both start from the same initial conditions as much as possible (both will start as hybrid initializations based on b20.007 0950-01-01). - There will still be some differences, as described in the 3 day validation tests. However, these cases will be much more closely related than the b21.002 and b20.007 runs namely in initialization and components. We are trying to isolate the cpl5/cpl6 differences to give further confidence to cpl6. These experiments will hopefully run 20 years each in about a week. If anyone would like to see the b21.002 and b20.007 runoff analysis, please stop by. I have hard copies of the relevant plots and am happy to go through them with anyone. Also, please feel free to share these results with anyone outside the CRB. I continue to gain confidence in cpl6. Hopefully, the final 2 cases will give others the same confidence. I still see no reason not to continue development of CCSM based on the cpl6 coupler. Thanks, tony.............. From tcraig@ucar.edu Fri Apr 4 11:27:45 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h34IRiaj024511 for ; Fri, 4 Apr 2003 11:27:45 -0700 Received: from sunray3.cgd.ucar.edu (sunray3.cgd.ucar.edu [128.117.24.203]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h34IPlCr016967 for ; Fri, 4 Apr 2003 11:25:47 -0700 (MST) Received: from localhost (tcraig@localhost) by sunray3.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h34IPkI12210 for ; Fri, 4 Apr 2003 11:25:46 -0700 (MST) X-Authentication-Warning: sunray3.cgd.ucar.edu: tcraig owned process doing -bs Date: Fri, 4 Apr 2003 11:25:46 -0700 (MST) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] Update X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 04 Apr 2003 18:27:45 -0000 CCSM-CRB, I just wanted to send a brief update on two items; cpl6 validation and changes in coupling fields. We think we've addressed all concerns with respect to cpl6 validation, and cpl6 has now been accepted by everyone that has reviewed results. That includes representatives from all model components. The final test was a 20 year round-off perturbation run with cpl6 that confirmed that differences in cpl5 and cpl6 20-year runs were consistent with roundoff differences. Bill Large's proposed changes to cross-component coupling are being implemented. In short, below is a list of what needs to change over the next couple weeks. Task Component Status ---- --------- ------ Modify Tref over ocean calculation coupler (done) Add Qref over ocean calculation coupler (done) Add UU_ref over ocean calculation coupler (done) Add Qref field to coupling list coupler (underway) recv from lnd/ice, send to atm Add UU_ref field to coupling list coupler (underway) send to ocean Merge Qref in coupler coupler (underway) Modify Tref over ice calculation ice (underway) Add Qref over ice calculation ice (underway) Add Qref over land calculation land (not started) Pack Qref into export fields land/ice (not started) Unpack Qref in import fields atm (not started) Unpack UU_ref in import fields ocean (not started) These changes are largely diagnostic and have no impact on our ability to carry out production runs. The coupler tasks should be done in the next couple days which will then allow the other tasks to be completed. If I've overlooked any task in these cross-component coupling changes, please let me know. Thanks, tony........ From tcraig@ucar.edu Tue Apr 8 15:30:26 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h38LUQaj025258 for ; Tue, 8 Apr 2003 15:30:26 -0600 Received: from sunray3.cgd.ucar.edu (sunray3.cgd.ucar.edu [128.117.24.203]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h38LS2Cr012339 for ; Tue, 8 Apr 2003 15:28:02 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray3.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h38LS4I14259 for ; Tue, 8 Apr 2003 15:28:04 -0600 (MDT) X-Authentication-Warning: sunray3.cgd.ucar.edu: tcraig owned process doing -bs Date: Tue, 8 Apr 2003 15:28:04 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] ice salinity, coupling fields update X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 08 Apr 2003 21:30:27 -0000 CCSM-CRB, We are going to change the ice reference salinty in the share constants from 0 to 4. The ice and ocean models have been validated with this value and we will verify this has no impact on the coupler budget diagnostics. If anyone has any concerns about this change, please let me know. Otherwise, we will proceed with this change in the next CCSM2.1 tag. The following line will change in shr_const_mod.F90, from real(SHR_KIND_R8),parameter :: SHR_CONST_ICE_REF_SAL = 0.0_SHR_KIND_R8 ! ice ref salinity (psu) to real(SHR_KIND_R8),parameter :: SHR_CONST_ICE_REF_SAL = 4.0_SHR_KIND_R8 ! ice ref salinity (psu) Also, just an update on Bill Large's proposed changes to cross-component coupling. Task Component Status ---- --------- ------ Modify Tref over ocean calculation coupler (done) Add Qref over ocean calculation coupler (done) Add UU_ref over ocean calculation coupler (done) Add Qref field to coupling list coupler (done) recv from lnd/ice, send to atm Add UU_ref field to coupling list coupler (done) send to ocean Merge Qref in coupler coupler (done) Modify Tref over ice calculation ice (underway) Add Qref over ice calculation ice (underway) Add Qref over land calculation land (proposed) Pack Qref into export fields land/ice (underway for ice) Unpack Qref in import fields atm (not started) Unpack UU_ref in import fields ocean (not started) Thanks, tony........ From tcraig@ucar.edu Tue Apr 15 10:22:24 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h3FGMOkn017494 for ; Tue, 15 Apr 2003 10:22:24 -0600 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.24.200]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h3FGMIJT005004 for ; Tue, 15 Apr 2003 10:22:18 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h3FGMNX09886 for ; Tue, 15 Apr 2003 10:22:23 -0600 (MDT) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Tue, 15 Apr 2003 10:22:23 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] Meeting: Wed, Apr 23 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 15 Apr 2003 16:22:24 -0000 CCSM-CRB, I'd like to propose a meeting next Wednesday, April 23, at 1pm to review the status of CCSM2.1. Two agenda items will be - modification of rain/snow partitioning in CCSM - proposed CCSM2.1 beta release Please suggest other agenda items. We decided to meet on Wed at 1pm (before CCSM-scientist meetings) at the last meeting, but if this meeting time is a problem, please let me know, and we'll shoot for a different time. Thanks, tony............. From tcraig@ucar.edu Wed Apr 16 12:39:42 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h3GIdgkn001425 for ; Wed, 16 Apr 2003 12:39:42 -0600 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.24.200]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h3GIdXJT021024 for ; Wed, 16 Apr 2003 12:39:33 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h3GIdda03653 for ; Wed, 16 Apr 2003 12:39:39 -0600 (MDT) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Wed, 16 Apr 2003 12:39:39 -0600 (MDT) From: Tony Craig X-Sender: To: Subject: [CCSM-CRB] Meeting: Wed, Apr 23 (fwd) Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Apr 2003 18:39:42 -0000 NOTE: This is next Wednesday. Thanks, tony........ ---------- Forwarded message ---------- Date: Tue, 15 Apr 2003 10:22:23 -0600 (MDT) From: Tony Craig To: ccsm-crb@cgd.ucar.edu Subject: [CCSM-CRB] Meeting: Wed, Apr 23 CCSM-CRB, I'd like to propose a meeting next Wednesday, April 23, at 1pm to review the status of CCSM2.1. Two agenda items will be - modification of rain/snow partitioning in CCSM - proposed CCSM2.1 beta release Please suggest other agenda items. We decided to meet on Wed at 1pm (before CCSM-scientist meetings) at the last meeting, but if this meeting time is a problem, please let me know, and we'll shoot for a different time. Thanks, tony............. _______________________________________________ CCSM-CRB mailing list CCSM-CRB@goldhill.cgd.ucar.edu http://www.cgd.ucar.edu/mailman/listinfo/ccsm-crb From tcraig@ucar.edu Tue Apr 22 11:44:46 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h3MHikkn001359 for ; Tue, 22 Apr 2003 11:44:46 -0600 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.24.200]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h3MHiMXJ010381 for ; Tue, 22 Apr 2003 11:44:22 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h3MHiS510104 for ; Tue, 22 Apr 2003 11:44:28 -0600 (MDT) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Tue, 22 Apr 2003 11:44:28 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] REMINDER- Meeting: Wed, Apr 23 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 22 Apr 2003 17:44:47 -0000 REMINDER: Tomorrow, 1pm, CRB meeting, CG1-2402 "the ice box". tony......... ---------- Forwarded message ---------- Date: Tue, 15 Apr 2003 10:22:23 -0600 (MDT) From: Tony Craig To: ccsm-crb@cgd.ucar.edu Subject: [CCSM-CRB] Meeting: Wed, Apr 23 CCSM-CRB, I'd like to propose a meeting next Wednesday, April 23, at 1pm to review the status of CCSM2.1. Two agenda items will be - modification of rain/snow partitioning in CCSM - proposed CCSM2.1 beta release Please suggest other agenda items. From tcraig@ucar.edu Wed Apr 23 11:31:48 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h3NHVmkn016433 for ; Wed, 23 Apr 2003 11:31:48 -0600 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.24.200]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h3NHVLXJ018108 for ; Wed, 23 Apr 2003 11:31:21 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h3NHVSo27923 for ; Wed, 23 Apr 2003 11:31:28 -0600 (MDT) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Wed, 23 Apr 2003 11:31:27 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CRB meeting agenda, April 23 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 23 Apr 2003 17:31:48 -0000 CRB meeting, Today, 1pm, CG1-2402 "the ice box". Agenda: - Plans for rain/snow partitioning - Proposed CCSM2.1 beta release - Tracking model changes during IPCC tuning - Plans for performance evalution - Naming conventions - Other tony......... From lydia@cgd.ucar.edu Tue May 13 08:01:49 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h4DE1nkn000494 for ; Tue, 13 May 2003 08:01:49 -0600 Received: from cgd.ucar.edu (dhcp42.cgd.ucar.edu [128.117.22.142]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h4DE0QYY018456 for ; Tue, 13 May 2003 08:00:26 -0600 (MDT) Message-ID: <3EC0FA8B.8050506@cgd.ucar.edu> Date: Tue, 13 May 2003 08:00:43 -0600 From: Lydia Shiver User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.1) Gecko/20020826 X-Accept-Language: en-us, en MIME-Version: 1.0 To: ccsm-crb@cgd.ucar.edu Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Tue, 13 May 2003 08:11:53 -0600 Subject: [CCSM-CRB] Notes on Meeting on Web X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 13 May 2003 14:01:50 -0000 Hi, I have updated the CRB notes on the web. Please see http://www.ccsm.ucar.edu/management/CCRB/ Thanks, Lydia From tcraig@ucar.edu Tue Jun 10 13:12:39 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h5AJCdOQ007242 for ; Tue, 10 Jun 2003 13:12:39 -0600 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.24.200]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h5AJBxFj011472 for ; Tue, 10 Jun 2003 13:11:59 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h5AJCQj10235 for ; Tue, 10 Jun 2003 13:12:26 -0600 (MDT) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Tue, 10 Jun 2003 13:12:26 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CCSM2.2 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 10 Jun 2003 19:12:39 -0000 CCSM-CRB, I'd like to propose the next CCSM internal tag be ccsm2_2_beta01. This would be the first ccsm2.2 tag and symbolic of the IPCC version. ccsm2.1 represents migration to cpl6. ccsm2.2 represents upgraded science/IPCC. For the time being, the end of year release would be called CCSM2.2, but that can change down the road if we decide we need a CCSM2.3 or CCSM3.0 name. I will proceed with this naming convention unless anyone has any concerns. Does anyone have any items to bring up within the CRB? Do we need to meet before Breckenridge to discuss any issues that might be proposed to the SSC or working groups? thanks, tony........... From tcraig@ucar.edu Tue Jun 17 11:32:19 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h5HHWJOQ011617 for ; Tue, 17 Jun 2003 11:32:19 -0600 Received: from sunray3.cgd.ucar.edu (sunray3.cgd.ucar.edu [128.117.21.23]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h5HHVT92005063 for ; Tue, 17 Jun 2003 11:31:29 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray3.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h5HHVwa16332 for ; Tue, 17 Jun 2003 11:31:58 -0600 (MDT) X-Authentication-Warning: sunray3.cgd.ucar.edu: tcraig owned process doing -bs Date: Tue, 17 Jun 2003 11:31:58 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] Meeting tomorrow (Wed, June 17) X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 17 Jun 2003 17:32:20 -0000 I'd like to propose a short CCSM-CRB meeting tomorrow at 1:30 in CG1-2414 to discuss - Tuning and monitoring of CCSM2.2 runs over the next few months - Support questions for CCSM2.0 and CCSM2.0.1 - Should we patch CCSM/CAM for the recently discovered CCSM exact restart bug - General questions about what "support" means. - Other items? I'm assuming there will be a CCSM scientists meeting at 2pm. thanks, tony............... From tcraig@ucar.edu Wed Jun 18 13:01:37 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h5IJ1bOQ027512 for ; Wed, 18 Jun 2003 13:01:37 -0600 Received: from sunray2.cgd.ucar.edu (sunray2.cgd.ucar.edu [128.117.21.22]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h5IJ0j92006700 for ; Wed, 18 Jun 2003 13:00:45 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray2.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h5IJ1FY03778 for ; Wed, 18 Jun 2003 13:01:15 -0600 (MDT) X-Authentication-Warning: sunray2.cgd.ucar.edu: tcraig owned process doing -bs Date: Wed, 18 Jun 2003 13:01:15 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] REMINDER Meeting today at 1:30 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 18 Jun 2003 19:01:37 -0000 ---------- Forwarded message ---------- Date: Tue, 17 Jun 2003 11:31:58 -0600 (MDT) From: Tony Craig To: ccsm-crb@cgd.ucar.edu Subject: [CCSM-CRB] Meeting tomorrow (Wed, June 17) I'd like to propose a short CCSM-CRB meeting tomorrow at 1:30 in CG1-2414 to discuss - Tuning and monitoring of CCSM2.2 runs over the next few months - Support questions for CCSM2.0 and CCSM2.0.1 - Should we patch CCSM/CAM for the recently discovered CCSM exact restart bug - General questions about what "support" means. - Other items - rain/snow partitioning From tcraig@ucar.edu Fri Jul 11 09:58:10 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h6BFw9BX017737 for ; Fri, 11 Jul 2003 09:58:09 -0600 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.21.21]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h6BFvRpn013509 for ; Fri, 11 Jul 2003 09:57:27 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h6BFw6q23586 for ; Fri, 11 Jul 2003 09:58:06 -0600 (MDT) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Fri, 11 Jul 2003 09:58:06 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CCSM CRB Meeting, July 21 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Jul 2003 15:58:11 -0000 I'd like to propose a CCSM-CRB meeting Monday, July 21 1:00-2:00 pm CG1-2414 The agenda includes - CCSM release plans - CCSM resolution and platform requirements - merging CAM and CCSM prescribed ice models Please let me know if you'd like to add other items. A CCSM scientist meeting will begin at 2:00pm. thanks, tony........... From tcraig@ucar.edu Mon Jul 21 09:20:55 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h6LFKsBX018206 for ; Mon, 21 Jul 2003 09:20:54 -0600 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.21.21]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h6LFJjpn010024 for ; Mon, 21 Jul 2003 09:19:45 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h6LFKSS01630 for ; Mon, 21 Jul 2003 09:20:28 -0600 (MDT) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Mon, 21 Jul 2003 09:20:28 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] Reminder: CCSM CRB Meeting, July 21 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Jul 2003 15:20:55 -0000 I'd like to propose a CCSM-CRB meeting Monday, July 21 1:00-2:00 pm CG1-2414 The agenda includes - CCSM release plans, release name - CCSM resolution and platform requirements (what resolutions and platforms will be support?) - recent ice and ocean fraction problems (related to roundoff to 0. or 1.) - new c2a So_t field - merging CAM and CCSM prescribed ice models - requirements for a CCSM single-executable model - bfb - som A CCSM scientist meeting will begin at 2:00pm. thanks, tony........... From lydia@cgd.ucar.edu Wed Jul 23 13:52:32 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h6NJqWBX019288 for ; Wed, 23 Jul 2003 13:52:32 -0600 Received: from cgd.ucar.edu (dhcp42.cgd.ucar.edu [128.117.22.142]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h6NJpJpn029333 for ; Wed, 23 Jul 2003 13:51:19 -0600 (MDT) Message-ID: <3F1EE761.3020804@cgd.ucar.edu> Date: Wed, 23 Jul 2003 13:52:01 -0600 From: Lydia Shiver User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.1) Gecko/20020826 X-Accept-Language: en-us, en MIME-Version: 1.0 To: ccsm-crb@goldhill.cgd.ucar.edu Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Wed, 23 Jul 2003 14:10:46 -0600 Subject: [CCSM-CRB] CRB Meeting Notes from 6/18/03 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 23 Jul 2003 19:52:32 -0000 Hello, The June 18th meeting notes are now on the web at: http://www.ccsm.ucar.edu/management/CCRB/mtgnotes/030618.html Thanks, Lydia From tcraig@ucar.edu Mon Aug 11 16:21:22 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h7BMLM8Y008551 for ; Mon, 11 Aug 2003 16:21:22 -0600 Received: from sunray3.cgd.ucar.edu (sunray3.cgd.ucar.edu [128.117.21.23]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h7BMKRoq028859 for ; Mon, 11 Aug 2003 16:20:27 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray3.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h7BMLI019423 for ; Mon, 11 Aug 2003 16:21:18 -0600 (MDT) X-Authentication-Warning: sunray3.cgd.ucar.edu: tcraig owned process doing -bs Date: Mon, 11 Aug 2003 16:21:18 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] Proposed Change to CCSM fractions X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 11 Aug 2003 22:21:22 -0000 CCSM-CRB, Below is a proposal for some changes to CCSM related to the fraction problems we've seen lately. Please review. I'd like this to be an agenda item at the next meeting, hopefully for both discussion and approval. Please suggests changes or ask for clarification. Thanks, tony.......... -------------------------------------------------- Proposed changes to "fractions" in CCSM Aug 11, 2003 "Fractions" - Fractions are the relative overlaps of grid boxes on the CCSM grids. Fractions have become a problem in CCSM from many perspectives. There is the issue of bounding them between [0,1], the fact that the sum of all overlaps should inherently add to 1.0, the loss of fractions of grid boxes because of filtering, the use of fractions as a mask, conservation of fractions through mapping, creation of mapped non-merged fields, and preservation of fractions as they are copied between arrays of different types. As long as this code exists in the coupler initialization, !--- NOTE: it's a requirement to elminate land points smaller than .001 --- !--- this is to avoid active land cells with tiny land fractions --- !--- NOTE: this may result in some, presumably small, non-conservation --- !--- NOTE: should probably count the number of pnts that get set to zero --- where (cpl_frac_a%data%rAttr(kl,:) < .001) cpl_frac_a%data%rAttr(kl,:) = 0.0_r8 there will be an inherent inability to conserve properties correctly to the 0.001 level for some grid boxes. This also has significant impacts on mapping of fractions and the requirement that fractions add to 1.0. Currently, the coupler handles fractions and o2a mapping as follows on ocn grid: fo = 1.0 - fi for "ocean points" fo = fi = 0.0 for "land points" on atm grid: masko -> o2a -> fl (initialization) where (fl < 0.001) fl = 0.0 (initialization) fi -> o2a -> fia (each hour) foa = 1.0 - fia - fl (each hour) mapping fields o2a: To*fo -> To' -> o2a -> Toa'/foa -> Toa foa can have relatively large errors if some combination of (fia ~ 1.0), (0.0 < fl << 1.0), and sum(mapping weights) <> 1.0. If this is the case, Toa will have some significant errors. For merged fields, this error is canceled. For unmerged fields, this error persists. Proposed Changes: - foa will be computed by mapping fo to the atm grid. It will no longer be computed as a residual. (fl + foa + fia) will no longer sum to 1.0. - the ice model will send fractions only in the range [0,1] - a new shr_const_fracerr = 1.0e-6 will be introduced. - for masking purposes, fractions that are 0.0+-shr_const_fracerr or 1.0+-shr_const_fracerr should be considered 0 or 1 by all components. - for computational purposes, fractions that are 0.0+-shr_const_fracerr or 1.0+-shr_const_fracerr can be considered 0.0 or 1.0 by all components. - the coupler will use shr_const_fracerr to set fractions to 0 or 1. the coupler will stop if a fraction less than -shr_const_fracerr or greater than 1.0+shr_const_fracerr is ever encountered. - the land model will use the shr_const_fracerr to test initial fractions received from the coupler against a pre-computed surface dataset. The following will continue: - mapping will be unchanged with small but acceptable and understood errors. - areas will still be used to correct fields going through the coupler. this corrects the discrepancy between scrip areas used to generate weights and areas reported by individual components. - fractions will continue to be used as part of the mapping process for o2a mappings because it produces significant improvements in the resulting fields from a scientific perspective. These changes will clarify the implementation of fractions and mapping and also correct problems encountered in both SE and science aspects of CCSM. From lydia@cgd.ucar.edu Mon Aug 11 16:09:44 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h7BM9h8Y008399 for ; Mon, 11 Aug 2003 16:09:43 -0600 Received: from cgd.ucar.edu (dhcp42.cgd.ucar.edu [128.117.22.142]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h7BM8moq028488 for ; Mon, 11 Aug 2003 16:08:48 -0600 (MDT) Message-ID: <3F381423.8070103@cgd.ucar.edu> Date: Mon, 11 Aug 2003 16:09:39 -0600 From: Lydia Shiver User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.1) Gecko/20020826 X-Accept-Language: en-us, en MIME-Version: 1.0 To: ccsm-crb@goldhill.cgd.ucar.edu Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Mon, 11 Aug 2003 16:25:28 -0600 Subject: [CCSM-CRB] CCRB Meeting NOte X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 11 Aug 2003 22:09:44 -0000 Hi, Here are the meetings notes from the July 21, 2003 meeting. http://www.ccsm.ucar.edu/management/CCRB/mtgnotes/030721.html Thanks, Lydia From tcraig@ucar.edu Fri Aug 15 09:14:08 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h7FFE88Y030666 for ; Fri, 15 Aug 2003 09:14:08 -0600 Received: from sunray3.cgd.ucar.edu (sunray3.cgd.ucar.edu [128.117.21.23]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h7FFD7oq002151 for ; Fri, 15 Aug 2003 09:13:07 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray3.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h7FFDwo21279 for ; Fri, 15 Aug 2003 09:13:58 -0600 (MDT) X-Authentication-Warning: sunray3.cgd.ucar.edu: tcraig owned process doing -bs Date: Fri, 15 Aug 2003 09:13:58 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] Proposed CLM bug-fix X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Aug 2003 15:14:09 -0000 CCSM-CRB, I just want to let everyone know that there will be a bug-fix applied to CLM in the next version of CCSM that will change answers but should not impact climate. If anyone has any concerns about this or wants to know more details, please let me know. Bill Collins is aware of the details associated with this change and believes it should go in as soon as possible. tony............ From tcraig@ucar.edu Tue Aug 26 10:38:06 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h7QGc62D010777 for ; Tue, 26 Aug 2003 10:38:06 -0600 Received: from sunray2.cgd.ucar.edu (sunray2.cgd.ucar.edu [128.117.21.22]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h7QGakgq014348 for ; Tue, 26 Aug 2003 10:36:46 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray2.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h7QGbh728360 for ; Tue, 26 Aug 2003 10:37:43 -0600 (MDT) X-Authentication-Warning: sunray2.cgd.ucar.edu: tcraig owned process doing -bs Date: Tue, 26 Aug 2003 10:37:42 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CCSM CRB meeting tomorrow X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Aug 2003 16:38:06 -0000 I'd like to propose a CCSM CRB meeting tomorrow, Wed, 8/27/03 1:00pm Room CG1-3416 The agenda includes - Discussion of CCSM release. Continued discussion of what features will be supported, hidden features, undocumented namelist, etc. - Discussion of CCSM fractions proposal. - Proposal that each CCSM component write a date stamp to its log file once per model day, with a format like "yyyy-mm-dd SSSSSs" - Proposal that each CCSM component write a consistent error-exit message when encountering a fatal error, with a format like "$COMPONENT stopping: $STRING" where $STRING is optional. - Proposal that each CCSM component write it's namelist to the log file and netcdf history file. - Problems with log files. The log files for nearly every component continue to have problems associated with multiple tasks writing to the same file and overwriting other print statements. If anyone has any other agenda items, please let me know. thanks, tony............ From tcraig@ucar.edu Wed Aug 27 11:00:33 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h7RH0X2D026237 for ; Wed, 27 Aug 2003 11:00:33 -0600 Received: from sunray2.cgd.ucar.edu (sunray2.cgd.ucar.edu [128.117.21.22]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h7RGxBgq022071 for ; Wed, 27 Aug 2003 10:59:11 -0600 (MDT) Received: from localhost (tcraig@localhost) by sunray2.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id h7RH07i15159 for ; Wed, 27 Aug 2003 11:00:07 -0600 (MDT) X-Authentication-Warning: sunray2.cgd.ucar.edu: tcraig owned process doing -bs Date: Wed, 27 Aug 2003 11:00:07 -0600 (MDT) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] Meeting Today X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 17:00:33 -0000 CCSM-CRB meeting today at 1pm, Room CG1-3416. Agenda and Notes follow: --------------------------------------------------- The agenda includes - Discussion of CCSM release. Continued discussion of what features will be supported, hidden features, undocumented namelist, etc. - Discussion of CCSM fractions proposal. - Proposal that each CCSM component write a date stamp to its log file once per model day, with a format like "yyyy-mm-dd SSSSSs" - Proposal that each CCSM component write a consistent error-exit message when encountering a fatal error, with a format like "$COMPONENT stopping: $STRING" where $STRING is optional. - Proposal that each CCSM component write it's namelist to the log file and netcdf history file. - Problems with log files. The log files for nearly every component continue to have problems associated with multiple tasks writing to the same file and overwriting other print statements. - Changes to common Makefile/Macro files which affect any component must be approved by that component's liaison prior to implementation. ------------------------------------------------------------------- Proposed changes to "fractions" in CCSM Aug 11, 2003 "Fractions" - Fractions are the relative overlaps of grid boxes on the CCSM grids. Fractions have become a problem in CCSM from many perspectives. There is the issue of bounding them between [0,1], the fact that the sum of all overlaps should inherently add to 1.0, the loss of fractions of grid boxes because of filtering, the use of fractions as a mask, conservation of fractions through mapping, creation of mapped non-merged fields, and preservation of fractions as they are copied between arrays of different types. As long as this code exists in the coupler initialization, !--- NOTE: it's a requirement to elminate land points smaller than .001 --- !--- this is to avoid active land cells with tiny land fractions --- !--- NOTE: this may result in some, presumably small, non-conservation --- !--- NOTE: should probably count the number of pnts that get set to zero --- where (cpl_frac_a%data%rAttr(kl,:) < .001) cpl_frac_a%data%rAttr(kl,:) = 0.0_r8 there will be an inherent inability to conserve properties correctly to the 0.001 level for some grid boxes. This also has significant impacts on mapping of fractions and the requirement that fractions add to 1.0. Currently, the coupler handles fractions and o2a mapping as follows on ocn grid: fo = 1.0 - fi for "ocean points" fo = fi = 0.0 for "land points" on atm grid: masko -> o2a -> fl (initialization) where (fl < 0.001) fl = 0.0 (initialization) fi -> o2a -> fia (each hour) foa = 1.0 - fia - fl (each hour) mapping fields o2a: To*fo -> To' -> o2a -> Toa'/foa -> Toa foa can have relatively large errors if some combination of (fia ~ 1.0), (0.0 < fl << 1.0), and sum(mapping weights) <> 1.0. If this is the case, Toa will have some significant errors. For merged fields, this error is canceled. For unmerged fields, this error persists. Proposed Changes: - foa will be computed by mapping fo to the atm grid. It will no longer be computed as a residual. (fl + foa + fia) will no longer sum to 1.0. - the ice model will send fractions only in the range [0,1] - a new shr_const_fracerr = 1.0e-6 will be introduced. - for masking purposes, fractions that are 0.0+-shr_const_fracerr or 1.0+-shr_const_fracerr should be considered 0 or 1 by all components. - for computational purposes, fractions that are 0.0+-shr_const_fracerr or 1.0+-shr_const_fracerr can be considered 0.0 or 1.0 by all components. - the coupler will use shr_const_fracerr to set fractions to 0 or 1. the coupler will stop if a fraction less than -shr_const_fracerr or greater than 1.0+shr_const_fracerr is ever encountered. - the land model will use the shr_const_fracerr to test initial fractions received from the coupler against a pre-computed surface dataset. The following will continue: - mapping will be unchanged with small but acceptable and understood errors. - areas will still be used to correct fields going through the coupler. this corrects the discrepancy between scrip areas used to generate weights and areas reported by individual components. - fractions will continue to be used as part of the mapping process for o2a mappings because it produces significant improvements in the resulting fields from a scientific perspective. These changes will clarify the implementation of fractions and mapping and also correct problems encountered in both SE and science aspects of CCSM. ------------------------------------------------------------------- - Proposal that each CCSM component write a date stamp to its log file once per model day, with a format like "yyyy-mm-dd SSSSSs" "Each active CCSM component model shall write a date stamp to its output log file once per model day. The print buffer should be "flushed" immediately following the write statement. A common date-stamp format is desired." Proposed modification: I like this idea. I'd like to see something of the form: cpl timestamp: yyyy-mm-dd Wall: yyyy-MM-dd hh:mm:ss dwall: ssssss awall: ssssss where cpl is the CCSM component name timestamp... is a string that can be grepped for followed by the current model date wall... is the current wall date and time dwall... is the difference in wall seconds between the previous printed timestamp and this one awall... is the average of dwall for this job. The wall/dwall combination lets us assess what else might be going on with the system when a slowdown is experienced. awall will give us a easy way to judge sustained performance Probably only the cpl needs to print dwall and awall It would be nice if there was a flush call immediately after this print, but I don'k know what the cost of that would be. ------------------------------------------------------------------- - Proposal that each CCSM component write a consistent error-exit message when encountering a fatal error, Proposal: ======== Each CCSM component model, when running in a CCSM coupled context and upon encountering a fatal error, must print a pre-defined CCSM error-exit message prior to model shut-down. The code for this could be available to all components via the "share" code. I propose the string "(component name) CCSM fatal error exit" or something to be determined by the CCSM CRB that conveys similar information. Background: ========== Presently, each component model is unrestricted in its choice of error-exit messages, and as one might expect from models that are developed somewhat or mostly indepently from one another, there are at least three different strings that are printed upon encountering an error condition. Benefits: ======== The implementation of this requirement would help to simplify the debugging of the CCSM. Instead of searching for each individual component model's error-exit string, the user would only need to search for the one common string. Costs: ===== The implementation of this requirement should require minimal effort on the part of the maintainers of the component model. *Suggested modification: "$COMPONENT fatal error exit" or something to signify fatal error and stopping. *Suggested addition: each CCSM component write a consistent successful-completion message with a format like "$COMPONENT successful completion" ------------------------------------------------------------------- From lydia@cgd.ucar.edu Fri Sep 5 12:56:30 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.5/8.12.5) with ESMTP id h85IuU2D003899 for ; Fri, 5 Sep 2003 12:56:30 -0600 Received: from cgd.ucar.edu (dhcp42.cgd.ucar.edu [128.117.22.142]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id h85It0gq022528 for ; Fri, 5 Sep 2003 12:55:00 -0600 (MDT) Message-ID: <3F58DC41.9070900@cgd.ucar.edu> Date: Fri, 05 Sep 2003 12:56:01 -0600 From: Lydia Shiver User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.1) Gecko/20020826 X-Accept-Language: en-us, en MIME-Version: 1.0 To: ccsm-crb@goldhill.cgd.ucar.edu Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Fri, 05 Sep 2003 12:59:08 -0600 Subject: [CCSM-CRB] CCRB Meeting Summary, 8/27/03 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 05 Sep 2003 18:56:30 -0000 CCSM-CRB meeting summary, Wed, 8/27/03 1:00pm Room CG1-3416 The agenda includes - Discussion of CCSM release. Continued discussion of what features will be supported, hidden features, undocumented namelist, etc. Bill and Gordon indicated that the Dust will need to be removed from CAM and CLM, and the Dynamic Vegetation will need to be removed from CLM. Bill encouraged others to come up with a list if other components have anything that needs to be removed before the model release. - Discussion of CCSM fractions proposal. See below for proposal. This proposal was deferred for now. Further study on the impact to each component will be made. A few additional suggestions were made. First, add a shr_const_lnderr value for use in the coupler instead of the hardwired (0.001). Second, should we really be thinking about using quantities instead of fluxes in the coupler. This would have an impact on overall coupler implementation including fractions. - Proposal that each CCSM component write a date stamp to its log file once per model day, with a format like "yyyy-mm-dd SSSSSs" This was accepted. Details of the requirement and the implementation will be determined. - Proposal that each CCSM component write a consistent error-exit message when encountering a fatal error, with a format like "$COMPONENT stopping: $STRING" where $STRING is optional. This was accepted. Details of the requirement and the implementation will be determined. - Proposal that each CCSM component write its namelist to the log file and netcdf history file. This was deferred for further exploration. In particular, the cost and impact of writing the namelist to netcdf files must be explored. There was general agreement that the namelist should be documented in model output somehow. - Problems with log files. The log files for nearly every component continue to have problems associated with multiple tasks writing to the same file and overwriting other print statements. This was deferred for future study. A idea was proposed that the log files be well defined and that new diagnostic files might also become a requirement. - Changes to common Makefile/Macro files that affect any component must be approved by that component's liaison prior to implementation. This was accepted. ------------------------------------------------------------------- Proposed changes to "fractions" in CCSM Aug 11, 2003 "Fractions" - Fractions are the relative overlaps of grid boxes on the CCSM grids. Fractions have become a problem in CCSM from many perspectives. There is the issue of bounding them between [0,1], the fact that the sum of all overlaps should inherently add to 1.0, the loss of fractions of grid boxes because of filtering, the use of fractions as a mask, conservation of fractions through mapping, creation of mapped non-merged fields, and preservation of fractions as they are copied between arrays of different types. As long as this code exists in the coupler initialization, !--- NOTE: it's a requirement to elminate land points smaller than .001 --- !--- this is to avoid active land cells with tiny land fractions --- !--- NOTE: this may result in some, presumably small, non-conservation --- !--- NOTE: should probably count the number of pnts that get set to zero --- where (cpl_frac_a%data%rAttr(kl,:) < .001) cpl_frac_a%data%rAttr(kl,:) = 0.0_r8 there will be an inherent inability to conserve properties correctly to the 0.001 level for some grid boxes. This also has significant impacts on mapping of fractions and the requirement that fractions add to 1.0. Currently, the coupler handles fractions and o2a mapping as follows: on ocn grid: fo = 1.0 - fi for "ocean points" fo = fi = 0.0 for "land points" on atm grid: masko -> o2a -> fl (initialization) where (fl < 0.001) fl = 0.0 (initialization) fi -> o2a -> fia (each hour) foa = 1.0 - fia - fl (each hour) mapping fields o2a: To*fo -> To' -> o2a -> Toa'/foa -> Toa foa can have relatively large errors if some combination of (fia ~ 1.0), (0.0 < fl << 1.0), and sum(mapping weights) <> 1.0. If this is the case, Toa will have some significant errors. For merged fields, this error is canceled. For unmerged fields, this error persists. Proposed Changes: - foa will be computed by mapping fo to the atm grid. It will no longer be computed as a residual. (fl + foa + fia) will no longer sum to 1.0. - the ice model will send fractions only in the range [0,1] - a new shr_const_fracerr = 1.0e-6 will be introduced. - for masking purposes, fractions that are 0.0+-shr_const_fracerr or 1.0+-shr_const_fracerr should be considered 0 or 1 by all components. - for computational purposes, fractions that are 0.0+-shr_const_fracerr or 1.0+-shr_const_fracerr can be considered 0.0 or 1.0 by all components. - the coupler will use shr_const_fracerr to set fractions to 0 or 1. The coupler will stop if a fraction less than -shr_const_fracerr or greater than 1.0+shr_const_fracerr is ever encountered. - the land model will use the shr_const_fracerr to test initial fractions received from the coupler against a pre-computed surface dataset. The following will continue: - mapping will be unchanged with small but acceptable and understood errors. - areas will still be used to correct fields going through the coupler. This corrects the discrepancy between scrip areas used to generate weights and areas reported by individual components. - fractions will continue to be used as part of the mapping process for o2a mappings because it produces significant improvements in the resulting fields from a scientific perspective. These changes will clarify the implementation of fractions and mapping and also correct problems encountered in both SE and science aspects of CCSM. ------------------------------------------------------------------- - Proposal that each CCSM component write a date stamp to its log file once per model day, with a format like "yyyy-mm-dd SSSSSs" "Each active CCSM component model shall write a date stamp to its output log file once per model day. The print buffer should be "flushed" immediately following the write statement. A common date-stamp format is desired." Proposed modification: I like this idea. I'd like to see something of the form: cpl timestamp: yyyy-mm-dd Wall: yyyy-MM-dd hh:mm:ss dwall: ssssss awall: ssssss where cpl is the CCSM component name timestamp... is a string that can be grepped for followed by the current model date wall... is the current wall date and time dwall... is the difference in wall seconds between the previous printed timestamp and this one awall... is the average of dwall for this job. The wall/dwall combination lets us assess what else might be going on with the system when a slowdown is experienced. awall will give us a easy way to judge sustained performance Probably only the cpl needs to print dwall and awall It would be nice if there was a flush call immediately after this print, but I don'k know what the cost of that would be. ------------------------------------------------------------------- - Proposal that each CCSM component write a consistent error-exit message when encountering a fatal error, Proposal: ======== Each CCSM component model, when running in a CCSM coupled context and upon encountering a fatal error, must print a pre-defined CCSM error-exit message prior to model shut-down. The code for this could be available to all components via the "share" code. I propose the string "(component name) CCSM fatal error exit" or something to be determined by the CCSM CRB that conveys similar information. Background: ========== Presently, each component model is unrestricted in its choice of error-exit messages, and as one might expect from models that are developed somewhat or mostly indepently from one another, there are at least three different strings that are printed upon encountering an error condition. Benefits: ======== The implementation of this requirement would help to simplify the debugging of the CCSM. Instead of searching for each individual component model's error-exit string, the user would only need to search for the one common string. Costs: ===== The implementation of this requirement should require minimal effort on the part of the maintainers of the component model. *Suggested modification: "$COMPONENT fatal error exit" or something to signify fatal error and stopping. *Suggested addition: each CCSM component write a consistent successful-completion message with a format like "$COMPONENT successful completion". From tcraig@ucar.edu Wed Dec 10 09:24:23 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBAGONa1008733 for ; Wed, 10 Dec 2003 09:24:23 -0700 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.21.21]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBAGMmtd017969 for ; Wed, 10 Dec 2003 09:22:48 -0700 (MST) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id hBAGOTV29437 for ; Wed, 10 Dec 2003 09:24:29 -0700 (MST) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Wed, 10 Dec 2003 09:24:29 -0700 (MST) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CCSM-CRB meeting, 12/15 10am X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 10 Dec 2003 16:24:23 -0000 Bill and I would like to have a CCSM CRB meeting on Monday, Dec 15 10:00 am Chapman Room The topic of discussion is the CCSM3.0 release - requirements - tasks - supported resolutions, platforms - control runs - schedule - code access prior to release A tentative schedule and task list is below. There is also a release checklist under development at http://www.cgd.ucar.edu/cseg/csm3.0/ccsm3.0check.txt It is not completely up-to-date right now, but I plan to update it this week and make it as current as I can. If the meeting time above is a problem, please let me know. tony............. ------------------------------------------------------ Release Plans: General Release Page Updated "I agree page" Source Code Single source code for all machines/resolutions Vector/SMP merged version New scripts / New GUI Target Platforms/Machines IBM - babyblue (NCAR), blackforest (NCAR), bluesky (NCAR), cheetah (ORNL), eagle (ORNL), seaborg (NERSC) SGI - chinook (NCAR), NEWguyot (LANL) HP/CPQ - lemieux (PSC) Cray X1 - phoenix (ORNL) NEC SX6 - ES?, rime (ARSC)? Linux - bangkok? (CGD), jazz (ANL) Resolutions and Controls all present day climate T85_gx1v3 (500 year control) T42_gx1v3 (1000 year control) T42_gx3v5 (?) T31_gx3v5 (1000 year control) 2x2.5_gx1v3 (?) T62_gx1v3 (?) latm run T62_gx3v5 (?) latm run Need to get gx3v5 and T31 versions of components ready Capability Exact restart demostrated for active models on T42_gx1v3 on all machines Harvest capability on all machines Batch script on all machines Input Datasets present day climate datasets, support above resolutions multiple tar files organized how? new latm, datm, dlnd datasets? Output Datasets pointers to control runs on mass store subsets of data on dataportal, what? Documentation users guides code developers guides science guides ------------------------------ Schedule: Controls: T85_gx1v3 (4 y/day, bluesky, 200 pes) T42_gx1v3 (6 y/day, blackforest, 200 pes) T31_gx3v5 (?, blackforest, ?) no later than dates: -------------------- Dec 15, 2003: Freeze CCSM3 physics Dec 15, 2003: Start T85_gx1v3, T42_gx1v3 controls Dec 19, 2003: Establish roundoff change CCSM3 climate test Jan 5, 2004: ES runs begin Jan 5, 2004: Establish inputdata datasets management strategy Jan 15, 2004: Start T31_gx3v5 control Feb 2, 2004: New scripts in production, drop old scripts Feb 2, 2004: Users guides for all component models ready Feb 2, 2004: Establish port test procedure and validation procedure Mar 1, 2004: All platforms are supported out of the box with new scripts Apr 1, 2004: New GUI in place and working Apr 1, 2004: Finalize model output release plan Apr 1, 2004: Code developers guides for all component models ready Apr 1, 2004: Finalize data and latm input datasets, if they are changing Apr 1, 2004: Establish formal model support strategy (patches, etc) Apr 1, 2004: Freeze model bfb, verify release produces same climate as control runs Apr 15, 2004: Users guide for CCSM3 ready May 1, 2004: T85_gx1v3 at year 500 T42_gx1v3 at year 1000 T31_gx1v3 at year 1000 May 1, 2004: Scientific guides for all component models ready May 6, 2004: Release data moved to dataportal May 10, 2004: Release page undergoes final testing, all release tasks complete May 18, 2004: Release available to the public From tcraig@ucar.edu Thu Dec 11 12:34:59 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBBJYxa1025581 for ; Thu, 11 Dec 2003 12:34:59 -0700 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.21.21]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBBJXPtd027967 for ; Thu, 11 Dec 2003 12:33:25 -0700 (MST) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id hBBJZ6911258 for ; Thu, 11 Dec 2003 12:35:06 -0700 (MST) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Thu, 11 Dec 2003 12:35:06 -0700 (MST) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CCSM-CRB meeting, 12/15 2:30pm X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 11 Dec 2003 19:35:00 -0000 NOTE: NEW MEETING TIME! There was a request to change the meeting time, see below --------------------------------------------------------- Bill and I would like to have a CCSM CRB meeting on Monday, Dec 15 2:30 pm Damon Room The topic of discussion is the CCSM3.0 release - requirements - tasks - supported resolutions, platforms - control runs - schedule - code access prior to release - future CCSM versions/releases A tentative schedule and task list is below. There is also a release checklist under development at http://www.cgd.ucar.edu/cseg/csm3.0/ccsm3.0check.txt It is not completely up-to-date right now, but I plan to update it this week and make it as current as I can. If the meeting time above is a problem, please let me know. tony............. ------------------------------------------------------ Release Plans: General Release Page Updated "I agree page" Source Code Single source code for all machines/resolutions Vector/SMP merged version New scripts / New GUI Target Platforms/Machines IBM - babyblue (NCAR), blackforest (NCAR), bluesky (NCAR), cheetah (ORNL), eagle (ORNL), seaborg (NERSC) SGI - chinook (NCAR), NEWguyot (LANL) HP/CPQ - lemieux (PSC) Cray X1 - phoenix (ORNL) NEC SX6 - ES?, rime (ARSC)? Linux - bangkok? (CGD), jazz (ANL) Resolutions and Controls all present day climate T85_gx1v3 (500 year control) T42_gx1v3 (1000 year control) T42_gx3v5 (?) T31_gx3v5 (1000 year control) 2x2.5_gx1v3 (?) T62_gx1v3 (?) latm run T62_gx3v5 (?) latm run Need to get gx3v5 and T31 versions of components ready Capability Exact restart demostrated for active models on T42_gx1v3 on all machines Harvest capability on all machines Batch script on all machines Input Datasets present day climate datasets, support above resolutions multiple tar files organized how? new latm, datm, dlnd datasets? Output Datasets pointers to control runs on mass store subsets of data on dataportal, what? Documentation users guides code developers guides science guides ------------------------------ Schedule: Controls: T85_gx1v3 (4 y/day, bluesky, 200 pes) T42_gx1v3 (6 y/day, blackforest, 200 pes) T31_gx3v5 (?, blackforest, ?) no later than dates: -------------------- Dec 15, 2003: Freeze CCSM3 physics Dec 15, 2003: Start T85_gx1v3, T42_gx1v3 controls Dec 19, 2003: Establish roundoff change CCSM3 climate test Jan 5, 2004: ES runs begin Jan 5, 2004: Establish inputdata datasets management strategy Jan 15, 2004: Start T31_gx3v5 control Feb 2, 2004: New scripts in production, drop old scripts Feb 2, 2004: Users guides for all component models ready Feb 2, 2004: Establish port test procedure and validation procedure Mar 1, 2004: All platforms are supported out of the box with new scripts Apr 1, 2004: New GUI in place and working Apr 1, 2004: Finalize model output release plan Apr 1, 2004: Code developers guides for all component models ready Apr 1, 2004: Finalize data and latm input datasets, if they are changing Apr 1, 2004: Establish formal model support strategy (patches, etc) Apr 1, 2004: Freeze model bfb, verify release produces same climate as control runs Apr 15, 2004: Users guide for CCSM3 ready May 1, 2004: T85_gx1v3 at year 500 T42_gx1v3 at year 1000 T31_gx1v3 at year 1000 May 1, 2004: Scientific guides for all component models ready May 6, 2004: Release data moved to dataportal May 10, 2004: Release page undergoes final testing, all release tasks complete May 18, 2004: Release available to the public _______________________________________________ CCSM-CRB mailing list CCSM-CRB@goldhill.cgd.ucar.edu http://www.cgd.ucar.edu/mailman/listinfo/ccsm-crb From tcraig@ucar.edu Mon Dec 15 16:14:10 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBFNEAGO001895 for ; Mon, 15 Dec 2003 16:14:10 -0700 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.21.21]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBFNCZJL020403 for ; Mon, 15 Dec 2003 16:12:35 -0700 (MST) Received: from localhost (tcraig@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id hBFNEId08659 for ; Mon, 15 Dec 2003 16:14:18 -0700 (MST) X-Authentication-Warning: sunray1.cgd.ucar.edu: tcraig owned process doing -bs Date: Mon, 15 Dec 2003 16:14:17 -0700 (MST) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CCSM-CRB meeting, 12/17, 9:30am X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 15 Dec 2003 23:14:10 -0000 We will continue the CCSM CRB meeting, Wed, Dec 17 9:30 am Damon Room We will continue discussion of the CCSM3.0 release I've included the notes in progress below. tony............. ------------------------------------------------------ Monday, Dec 15 2:30 pm Damon Room The topic of discussion is the CCSM3.0 release - requirements - tasks - supported resolutions, platforms - control runs - schedule - code access prior to release - future CCSM versions/releases ------------------------------------------------ General Requirements - Single Source code for Vector/SMP, machines/resolutions - New Scripts/GUI - Documentation; users, code developers, science - Update "I Agree" Download Page - F ice model, prescribed snow? - Backward compatability of restart files Resolutions, Controls: gx3v5 gx1v3 ----- ----- T31 B-1000 X T42 X B-1000 T62 G,M,L? G,M,L? T85 X B-500 2x2.5 X B-? T31 tuning schedule, lead? Jim Hack, next week or two gx3v5 tuning schedule, lead? ice model initial condition Nancy Norton Data model datasets? docn/dice - generate AMIP SST gx3v5 dataset? support gx3v5 and gx1v3 datm - new (valid) F run, T42, T31, T85 dlnd - same at datm latm - T62, "out of box", talk to Bill Large, M and G are different, community usage?, dataset release?, dataset documentation Freeze/Roundoff/bfb/Controls Plans? Portion of run needs to be reproducible Continue controls until March 1 or complete Last 100+ years of controls to be bfb with release code, bfb freeze no later than April 1 Climate equivalence test? Bill is leading Use surface temperature Should work at different resolutions Shoot for less than 10 years run integration Model Output from Controls, what, where, how? Platforms: bluesky (IBM SP4 NCAR ) blackforest (IBM SP3 NCAR ) cheetah (IBM SP4 ORNL ) eagle (IBM SP3 ORNL ) seaborg (IBM SP3 NERSC ) chinook (SGI O3K NCAR ) ? (SGI Altix LANL ) jazz (Linux ANL ) bangkok (Linux NCAR/CGD) ? (Linux Bern ) lemieux (CPQ PSC ) phoenix (Cray X1 ORNL ) moon (NEC ES ) rime (NEC ARSC ) Code access prior to release? Future Release Issues, bgc, etc? Schedule: no later than dates: -------------------- Dec 15, 2003: Freeze CCSM3 physics Dec 15, 2003: Start T85_gx1v3, T42_gx1v3 controls Dec 19, 2003: Establish roundoff change CCSM3 climate test Jan 5, 2004: ES runs begin Jan 5, 2004: Establish inputdata datasets management strategy Jan 15, 2004: Start T31_gx3v5 control Feb 2, 2004: New scripts in production, drop old scripts Feb 2, 2004: Users guides for all component models ready Feb 2, 2004: Establish port test procedure and validation procedure Mar 1, 2004: All platforms are supported out of the box with new scripts Apr 1, 2004: New GUI in place and working Apr 1, 2004: Finalize model output release plan Apr 1, 2004: Code developers guides for all component models ready Apr 1, 2004: Finalize data and latm input datasets, if they are changing Apr 1, 2004: Establish formal model support strategy (patches, etc) Apr 1, 2004: Freeze model bfb, verify release produces same climate as control runs Apr 15, 2004: Users guide for CCSM3 ready May 1, 2004: T85_gx1v3 at year 500 T42_gx1v3 at year 1000 T31_gx3v5 at year 1000 May 1, 2004: Scientific guides for all component models ready May 6, 2004: Release data moved to dataportal May 10, 2004: Release page undergoes final testing, all release tasks complete May 18, 2004: Release available to the public From tcraig@ucar.edu Tue Dec 16 12:16:38 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBGJGcGO013268 for ; Tue, 16 Dec 2003 12:16:38 -0700 Received: from sunray2.cgd.ucar.edu (sunray2.cgd.ucar.edu [128.117.21.22]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBGJF3JL028235 for ; Tue, 16 Dec 2003 12:15:03 -0700 (MST) Received: from localhost (tcraig@localhost) by sunray2.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id hBGJHE007588 for ; Tue, 16 Dec 2003 12:17:14 -0700 (MST) X-Authentication-Warning: sunray2.cgd.ucar.edu: tcraig owned process doing -bs Date: Tue, 16 Dec 2003 12:17:14 -0700 (MST) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CCSM-CRB meeting, 12/17, 10:00am X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 16 Dec 2003 19:16:38 -0000 The CCSM-CRB meeting start time has been pushed back 30 minutes to 10am. See below. thanks, tony........... ------------------------------------------------------- We will continue the CCSM CRB meeting, Wed, Dec 17 10am Damon Room We will continue discussion of the CCSM3.0 release I've included the notes in progress below. tony............. ------------------------------------------------------ Monday, Dec 15 2:30 pm Damon Room The topic of discussion is the CCSM3.0 release - requirements - tasks - supported resolutions, platforms - control runs - schedule - code access prior to release - future CCSM versions/releases ------------------------------------------------ General Requirements - Single Source code for Vector/SMP, machines/resolutions - New Scripts/GUI - Documentation; users, code developers, science - Update "I Agree" Download Page - F ice model, prescribed snow? - Backward compatability of restart files Resolutions, Controls: gx3v5 gx1v3 ----- ----- T31 B-1000 X T42 X B-1000 T62 G,M,L? G,M,L? T85 X B-500 2x2.5 X B-? T31 tuning schedule, lead? Jim Hack, next week or two gx3v5 tuning schedule, lead? ice model initial condition Nancy Norton Data model datasets? docn/dice - generate AMIP SST gx3v5 dataset? support gx3v5 and gx1v3 datm - new (valid) F run, T42, T31, T85 dlnd - same at datm latm - T62, "out of box", talk to Bill Large, M and G are different, community usage?, dataset release?, dataset documentation Freeze/Roundoff/bfb/Controls Plans? Portion of run needs to be reproducible Continue controls until March 1 or complete Last 100+ years of controls to be bfb with release code, bfb freeze no later than April 1 Climate equivalence test? Bill is leading Use surface temperature Should work at different resolutions Shoot for less than 10 years run integration Model Output from Controls, what, where, how? Platforms: bluesky (IBM SP4 NCAR ) blackforest (IBM SP3 NCAR ) cheetah (IBM SP4 ORNL ) eagle (IBM SP3 ORNL ) seaborg (IBM SP3 NERSC ) chinook (SGI O3K NCAR ) ? (SGI Altix LANL ) jazz (Linux ANL ) bangkok (Linux NCAR/CGD) ? (Linux Bern ) lemieux (CPQ PSC ) phoenix (Cray X1 ORNL ) moon (NEC ES ) rime (NEC ARSC ) Code access prior to release? Future Release Issues, bgc, etc? Schedule: no later than dates: -------------------- Dec 15, 2003: Freeze CCSM3 physics Dec 15, 2003: Start T85_gx1v3, T42_gx1v3 controls Dec 19, 2003: Establish roundoff change CCSM3 climate test Jan 5, 2004: ES runs begin Jan 5, 2004: Establish inputdata datasets management strategy Jan 15, 2004: Start T31_gx3v5 control Feb 2, 2004: New scripts in production, drop old scripts Feb 2, 2004: Users guides for all component models ready Feb 2, 2004: Establish port test procedure and validation procedure Mar 1, 2004: All platforms are supported out of the box with new scripts Apr 1, 2004: New GUI in place and working Apr 1, 2004: Finalize model output release plan Apr 1, 2004: Code developers guides for all component models ready Apr 1, 2004: Finalize data and latm input datasets, if they are changing Apr 1, 2004: Establish formal model support strategy (patches, etc) Apr 1, 2004: Freeze model bfb, verify release produces same climate as control runs Apr 15, 2004: Users guide for CCSM3 ready May 1, 2004: T85_gx1v3 at year 500 T42_gx1v3 at year 1000 T31_gx3v5 at year 1000 May 1, 2004: Scientific guides for all component models ready May 6, 2004: Release data moved to dataportal May 10, 2004: Release page undergoes final testing, all release tasks complete May 18, 2004: Release available to the public _______________________________________________ CCSM-CRB mailing list CCSM-CRB@goldhill.cgd.ucar.edu http://www.cgd.ucar.edu/mailman/listinfo/ccsm-crb From tcraig@ucar.edu Wed Dec 17 12:38:08 2003 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBHJc8GO027511 for ; Wed, 17 Dec 2003 12:38:08 -0700 Received: from sunray3.cgd.ucar.edu (sunray3.cgd.ucar.edu [128.117.21.23]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id hBHJaYJL015041 for ; Wed, 17 Dec 2003 12:36:34 -0700 (MST) Received: from localhost (tcraig@localhost) by sunray3.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id hBHJbCe01028 for ; Wed, 17 Dec 2003 12:37:12 -0700 (MST) X-Authentication-Warning: sunray3.cgd.ucar.edu: tcraig owned process doing -bs Date: Wed, 17 Dec 2003 12:37:12 -0700 (MST) From: Tony Craig X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CCSM3.0 Release CRB Notes and Checklist X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 17 Dec 2003 19:38:08 -0000 CCSM-CRB, I will ask Lydia to put the CCSM-CRB notes on the official CCSM web site after she's back. For now, the CRB notes can be found off the CSEG page, http://www.cgd.ucar.edu/cseg/ near the top. login/password are the usual. Direct links are CRB Notes- http://www.cgd.ucar.edu/cseg/csm3.0/ccsm_crb_release30_notes.txt Release Checklist- http://www.cgd.ucar.edu/cseg/csm3.0/ccsm3.0check.txt The release checklist will evolve over time. Consider this a first draft. I will try to update it over the next few weeks to include the schedule and some "green/red" check boxes. Thanks for the time and contributions from everyone. This meeting was critical for getting the release organized. tony......... ---------- Forwarded message ---------- Date: Tue, 16 Dec 2003 12:17:14 -0700 (MST) From: Tony Craig To: ccsm-crb@cgd.ucar.edu Subject: [CCSM-CRB] CCSM-CRB meeting, 12/17, 10:00am Monday, Dec 15 2:30 pm Damon Room The topic of discussion is the CCSM3.0 release - requirements - tasks - supported resolutions, platforms - control runs - schedule - code access prior to release - future CCSM versions/releases From tcraig@ucar.edu Tue Feb 10 09:21:44 2004 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id i1AGLi4O005631; Tue, 10 Feb 2004 09:21:44 -0700 Received: from sunray3.cgd.ucar.edu (sunray3.cgd.ucar.edu [128.117.21.23]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id i1AGKOHE006750; Tue, 10 Feb 2004 09:20:24 -0700 (MST) Received: from localhost (tcraig@localhost)i1AGKXZ06546; Tue, 10 Feb 2004 09:20:33 -0700 (MST) X-Authentication-Warning: sunray3.cgd.ucar.edu: tcraig owned process doing -bs Date: Tue, 10 Feb 2004 09:20:33 -0700 (MST) From: Tony Craig X-Sender: To: , Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [CCSM-CRB] CCSM3 Release Date, June 2, 2004 X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 10 Feb 2004 16:21:44 -0000 FYI and in case you haven't heard, the release date for CCSM3 has been pushed back to June 2. Schedules and milestones for the release will be updated to reflect this new date. I do not expect the release date to be changed again unless there are extreme circumstances largely outside our control. This change gives us an extra 2 weeks to put effort into the release to improve quality and still allows us to have the release out a month before the summer meeting. Also, this still meets the "release by" date set by the SSC. Brian Kauffman will be coordinating preparation of the release. There is a CCSM3 release plan linked off the CSEG page, www.cgd.ucar.edu/cseg. CSEG will be spending most of their time over the next few months helping finalize model configurations, testing, documenting, doing control runs, preparing model input and output datasets, and getting the web site ready. If you have any questions, please feel free to contact Brian or myself. tony.......... From kauff@cgd.ucar.edu Thu Feb 26 13:23:25 2004 Received: from tablemtn.cgd.ucar.edu (tablemtn.cgd.ucar.edu [128.117.21.12]) by echorock.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id i1QKNP5W016359 for ; Thu, 26 Feb 2004 13:23:25 -0700 Received: from sunray1.cgd.ucar.edu (sunray1.cgd.ucar.edu [128.117.21.21]) by tablemtn.cgd.ucar.edu (8.12.8/8.12.8) with ESMTP id i1QKL2HE008750 for ; Thu, 26 Feb 2004 13:21:02 -0700 (MST) Received: from localhost (kauff@localhost) by sunray1.cgd.ucar.edu (8.11.6+Sun/8.11.6) with ESMTP id i1QKNEe21741 for ; Thu, 26 Feb 2004 13:23:14 -0700 (MST) X-Authentication-Warning: sunray1.cgd.ucar.edu: kauff owned process doing -bs Date: Thu, 26 Feb 2004 13:23:14 -0700 (MST) From: Brian Kauffman X-Sender: To: In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Mailman-Approved-At: Thu, 26 Feb 2004 15:35:59 -0700 Subject: [CCSM-CRB] ccsm3.0 release - code freeze dates X-BeenThere: ccsm-crb@goldhill.cgd.ucar.edu X-Mailman-Version: 2.1 Precedence: list List-Id: CCSM Change Review Board List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 26 Feb 2004 20:23:25 -0000 CCSM-CRB, CSEG has put together a CCSM3.0 release timetable which is available at http://www.cgd.ucar.edu/cseg/csm3.0/checklist/checklist_milestones.html It remains very similar to the time table presented at the 2003 Dec 15 CCRB meeting, the notes of which are available at http://www.ccsm.ucar.edu/management/CCRB There are a few dates of particular importance... Mar 15 create a CCSM tag with unified vector/scalar code Apr 1 FREEZE CCSM3: BFB (on IBM wrt 1990 controls ~ IPCC mode OFF) that means all subsequent code changes must be bit-for-bit Apr 30 FREEZE CCSM3: SRC CODE AND DATA SETS that's the cut off date for all changes to source code, this allows one month of final testing, bug fixes, and preparation for release Jun 2 release available to the public If anyone feels this schedule is not OK, please raise the issue ASAP with Bill Collins and/or the CCRB. thanks, -Brian