WRF-DFI project meeting – 20060606

Hans, Wei, Min, Ju-Won

Lilly room, 2-3pm, Tuesday 6 Jun 2006

1.Work review

1.1. Statistics

The divergence tendency field and the domain-averaged precipitation time series are still strange on the CONUS case, too. Hans suggested plotting the divergence tendency field at model level. Wei suggested making the test on the CONUS domain with 90km resolution on bluesky. She will check the code related to the divergence and precipitation fields. She had comments on MU field. There is something wrong in MU1. Min had comment on the precipitation oscillations. She has also seen the oscillations in MM5 results.

Hans asked Ju-Won to start drafting a document of the statistics with formulae, description, and their possible use in the DFI related studies.

1.2. DFL

Min should repeat some of her DFL experiments on bluesky.

Ju-Won should run more long time for the DFL cycle tests, using the KMA 10km domain.

A documentation of DFL is still needed.

1.3. Backward integration

Tom Henderson updated the ESMF timer so that now the timer will go backward properly with negative integration time and negative time step. The alarm is also working which means output can be written. The updated code is in the latest repository.

Wei started the backward integration work using the em_b_wave case, and updated code from Tom. The idea is to work on the dynamics for backward integration first without worrying about the lateral boundary condition. An initial test of the backward integration shows that the model runs, with correct backward time stepping and IO. But Wei will check the correctness of the results, and report back to the group at the next meeting.

2.Work plan for two weeks

All: try to understand the strange oscillations in some statistics.

Wei: press on the backward integration work using the idealized cases. Check the code status to see if we should bring our code to the top of the repository. One reason for doing this would be to incorporate the latest modification to ESMF timer from Tom.

Min: get familiar with the NCAR computing environment; repeat the DFL tests; write some text about the DFL.

Ju-Won: run the DFL experiments (cold start and cycle tests), ask for the serial case data to the KMA about 1 month, repeat the DFL tests on the CONUS domain with 90km resolution and confirm the noisy pattern in the divergence tendency fields and the domain-averaged quantities; write some text about the statistics package.

Hans: help me with practical arrangement; respond to FSL if necessary; review the statistics document and DFL document.

3.AOB

Hans had comments on CVS controls. Although it is convenient to use CVS to bring new changes to the reference system, it is not worse the trouble for Min and Ju-Won to learn CVS as it is already replaced by svn.

Due to the WRF workshop, we will move the next meeting to: Thursday 15 June, 2 pm, Lilly room.