Adele Morrison Kial Stewart Andy Hogg Paul Spence Aidan Heerdegen S econd global reanalysis conducted by the Japan Meteorological Agency Utilises fourdimensional variational ID: 638088
Download Presentation The PPT/PDF document "Switching from CORE to JRA-55 forcing" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Switching from CORE to JRA-55 forcing
Adele Morrison,
Kial
Stewart, Andy Hogg, Paul Spence, Aidan
Heerdegen
Slide2
Second global reanalysis
conducted by the Japan Meteorological Agency.
Utilises
four-dimensional variational analysis (4D-Var).Generally improved representations of phenomena on wide range of space-time scales.Endorsed by CLIVAR Ocean Model Development Panel, used by other groups (e.g. NCAR).1958-2012 (55 years) + ongoing (2013-present).0.5° resolution, 3-hourly.
Why JRA-55?
Frequency of spatial correlation of daily
precip
with obs over tropics (22°S-22°N)
Harada et al. 2016Slide3
The second global reanalysis conducted
by the Japan Meteorological Agency.
Utilises
four-dimensional variational analysis (4D-Var).Generally improved representations of phenomena on wide range of space-time scales.Endorsed by CLIVAR Ocean Model Development Panel, used by other groups (e.g. NCAR).1958-2012 (55 years) + ongoing (2013-present).0.5° resolution, 3-hourly
Why JRA-55?
Surface air temperature trend (K/
dec
), 1958-2001, compared with obs (HadCRUT4) with 80% data availability.
RMSE 0.099
RMSE 0.151
RMSE 0.129
Harada et al. 2016Slide4
Bias adjustment needed for ocean/sea-ice modelling
(as done for CORE and DRAKKAR projects) - led by Hiroyuki Tsujino (JMA-MRI) as part of OMDP-JRA55 collaborative effort.
Version
1.1 recently available including runoff; documentation will be complete by end of year.Intended for use in OMIP – version 2 of CMIP6.Ocean forcing dataset from JRA-55Slide5
CORE used a Normal Year Forcing (NYF) produced by a spectral averaging technique, but had many drawbacks (spatially noisy, lacking weather variance
etc).Proposal is to replace with RYF – repeated single year forcing.
Years are cut from 1 May to 30 April to avoid large discontinuities mid-season.
Selection of Repeat Year Forcing (RYF)
JRA-55
Our
uber
index = 4*SOI + 2*SAM + 1*NAO
Proposed years for testing: 1984-85, 1990-91, 2003-04.
NCAR decided on same 3 years using
uber
index = 1*NAO + 1*SOI + 0.8*SAM + 0.5*NPISlide6
We have yet to test in our models – will begin
asap.Comparison of RYF years in 1° NCAR model:
Selection of Repeat Year Forcing (RYF)
AMOC
NCAR prefers 1990-91 RYF, due to reduced global cooling, larger AMOC.
Suggested that different groups may select different RYF years, due to strong sensitivity of model climate state to model choices and salinity restoring.
NCAR is awaiting our RYF evaluation before making public recommendations and data releases of RYF.
What diagnostics do we care about?Slide7
Initial conditions –
WOA13v2 temperature and salinity averaged over 1955-2012, – 0.25°, January in top 1500 m, ‘winter’ average below 1500 m (same as NCAR).
WOA13v2 data used for sea surface salinity restoring
– 0.25°, average of 1.25 m and 10 m levels (following NCAR protocol). Will test restoring timescale / salinity cap for JRA-55 forced models. Other changes to model configurationSlide8
Initial conditions –
WOA13v2 temperature and salinity averaged over 1955-2012, – 0.25°, January in top 1500 m, ‘winter’ average below 1500 m (same as NCAR).
WOA13v2 data used for sea surface salinity restoring
– 0.25°, average of 1.25 m and 10 m levels (following NCAR protocol). Will test restoring timescale / salinity cap for JRA-55 forced models. Splitting of JRA-55 Antarctic runoff into basal melt and iceberg melt (distributed spatially according to Merino et al. 2016 climatology).Other changes to model configuration