If you are using
Navigator 4.x
or
Internet Explorer 4.x
or
Omni Web 4.x
, this site will not render
correctly!
gfdl's home page > people > John Dunne >
Notes from May 24, 2006
The purpose of these meetings was to present a preliminary analysis of our current 1860 run, and more clearly define the guiding philosophy for ESM development and implementation through a discussion of the various potential physical initialization strategies, carbon initialization strategies and potential carbon system "show stoppers".
1) Ron Stouffer is a new ESMDT leader, replacing Jorge Sarmiento
2. Proto-type ESM model run is out 200+ years (John D.)
- Two experiments are currently registered on the database:
ESM2p1U_co2_ov_lm2_test_bgc which has interactive ocean
biogeochemistry from a memphis bgc version, and
ESM2p1U_co2_ov_lm3_v1 which has memphis LM3v and a slightly more
advanced ocean bgc. Both were initialized from
/archive/ccsp/ipcc_ar4/CM2.1U_Control-1860_D4/restart/19930101.cpio
and can be compared with
/archive/ccsp/ipcc_ar4/CM2.1U_Control-1860_D4/pp. Unfortunately, most atmospheric postprocessing
data is currently missing due to unresolved
problems.
- The ENSO has changed character from earlier test
integrations to become much more bi-annual in nature without
longer periodicity. The
reasons for the changes are unclear.
- New code was included into the model to add river runoff of
DIC and alkalinity, compute the CaCO3 saturation
depth, and allow for permanent CaCO3 burial. Analysis shows:
The CaCO3 saturation depth simulation seems reasonable. Problems near
continents in tropics are noted.
- New code was included to flux tracers in rivers to the oceans. The
impact on the simulation was relatively small because the model is not
nitrogen limited. The ocean chemistry needs retuned to reduce the nitrogen
in the surface waters. These efforts are part of a demonstration project
for a NOAA milestone.
- John has been doing experiments to retune the obgc after switching "observed" atmospheric iron deposition fields from
Ginoux et al., 2001 to the new Fan et al. scheme. The new
fields increase the iron supply to the ocean considerably.
- His presentation can be downloaded
here
3. Land vegetation - Putting LM3V into CM2.1 (ESM2.1 - see #2 above) (Elena S.)
- Initialized land model using AM2 temperatures and observed
precipitation
Western US becomes forested compared to initial conditions, but
overall simulation "not too bad"
- After 100 years NPP becomes fairly constant, assessment is
that:
- model has too much fire (big fire spike in first few years
due to initialization problem of intense first-year El Nino)
- Respiration not steady after 200 years
- NH continents seem to be responsible for most of NPP drift
- Water supply changes seem to be cause of NPP drifts
- Issues
- Tune land model or physical climate model to improve
precipitation
- How to initialize land to 1860? And more generally what is the
initialization scheme for the land/ocean.
- Koppen as example - 5 major climate types
- Based on observations with distinct min/max temperature threshholds
- Model looks pretty good overall, details not so good
- HadGEM looks very good relative to GFDL and NCAR, yet temperature and precipitation errors are larger in
HadGEM, re-enforcing notion of the filtering of the classification
scheme
- Potentially a good metric for ESM development
- Suggests how good one should expect the interactive/predicted
vegetation will be when included in the model - i.e. that there are a
quite a number of terrestrial ecology misfits that should be
expected given misfits in climate alone, independent of the
model's ecological representation.
- His
presentation can be downloaded
here