Evaluation of Tropical Pacific Observing Systems Using NCEP and GFDL Ocean Data Assimilation Systems
Y. Xue1, C. Wen1, X. Yang2 , D. Behringer1, A. Kumar1,
G. Vecchi2, A. Rosati2, R. Gudgel2
1 NCEP/NOAA, USA
2 GFDL/NOAA, USA
GODAE OceanView Observing System Evaluation Task Team and GSOP-CLIVAR Workshop, Toulouse, France, Dec 11-12, 2014
Seasonal to Interannual Forecasting at NCEP
Global Ocean
Data Assimilation
System (GODAS)
Climate Forecast
System (CFSv1)
SST XBT
Moorings Altimeter
Argo
R2 Surface Fluxes
SST Anomaly Forecast Forecasters
Official ENSO Forecast
Official Probabilistic
Surface Temperature
& Rainfall Forecasts
Seasonal Forecasts
for North America
CCA, CA
Markov
CCA, OCN
MR, ENSO
Ocean Initial Conditions
IRI
IRI
Global Ocean Monitoring Products Based on GODAS (A partnership between CPC and COD/CPO to deliver climate relevant products to the society)
http://www.cpc.ncep.noaa.gov/products/GODAS
• Synthesis of global ocean observations by NCEP’s Global Ocean Data Assimilation System (GODAS)
• Monthly Ocean Briefing
• Products used widely by operational climate prediction centers, researchers, fishery managers, news media, program managers, teachers and students
Contact: Yan Xue, NOAA/CPC
Synthesis of Ocean Observations
Ocean Observations from Saha et al. (2010)
Argo
(after 2001)
Altimetry
(after 1993)
XBT
Argo
TAO
(after 1985)
TAO
XBT
1985
1995 minus 1985
2006 minus 1995
Ensemble Spread Data Count Intercomparison of
Upper 300 Heat Content Xue et al. 2011, J Clim
6
Tropical Pacific Observing Systems
Black: All data Red: TAO/TRITON Blue: XBT Green: Argo
3S-3N
8S-8N
2004-2011
7
CTL ALL noMoor noArgo
In situ data
included
no profiles
OISST
all profiles
OISST
all except moorings
OISST
all except Argo
OISST
XBT × √ √ √
TAO × √ × √
Argo × √ √ ×
Coordinated Observing System Experiments At NCEP and GFDL (2004-2011)
What are the mean biases and RMSE (monthly anomalies) in the control simulations when no in situ observations are assimilated? How well are mean biases and RMSE constrained by assimilation of all in situ observations? What are influences of withholding mooring or Argo data on mean biases and RMSE?
NCEP’s Global Ocean Data Assimilation System
• MOM4p1, 0.5o resolution, 1/4o in 10oS-10oN, 40 levels
• Daily fluxes from NCEP Reanalysis 2
• 3D-VAR, Univariate in temperature and salinity (Behringer 2007)
• Horizonal covariance with scale elongated in zonal direction, vertical covariance as function of local vertical temperature gradient
• Temperature profiles and OISST assimilated
• Synthetic salinity derived with climatology T/S relationship and temperature profiles from XBT and moorings, observed salinity from Argo assimilated
• Altimetry SSH not assimilated
• Temperature (salinity) at 5m is relaxed to daily OISST (WOA salinity climatology) with 10 (30) day relaxation time scales.
GFDL’s Ensemble Coupled Data Assimilation
• CM2.1 coupled climate model
• MOM4 with 1o resolution, 1/3o near equator, 50 levels
• Ensemble Kalman filter with 12-ensemble members
• Atmosphere is constrained by NCEP reanalyses (3D air temperature, u, v assimilated)
• Temperature profiles and OISST assimilated
• Pseudo salinity derived with vertical coupled T-S EOFs and temperature profiles from XBT and moorings and altimetry SSH, observed salinity from Argo assimilated
• Altimetry SSH not assimilated
• Ensemble mean is analyzed
Evaluation of OSEs
• Evaluation data
– TAO temperature and current
– Altimetry SSH from AVISO
– Surface current from OSCAR
– EN4 temperature and salinity analysis (objective analysis based on in situ data only) is used as reference but not the “truth”
• Evaluation Methods
– Mean and annual cycle
– Standard deviation (STD)
– Root-mean-square error (RMSE) and anomaly correlation coefficient (ACC)
– Integrated RMSE in upper 300m
11
Mean Biases in Upper 300m Heat Content
12
Mean Biases in Temperature Averaged in 2S-2N
13
STD Biases of Temperature Averaged in 2S-2N
14
TAO (dotted line) ALL noMoor noArgo CTL
Temperature Averaged in 130W-100W, 2S-2N at z=75m
15
Mean Biases of Temperature Averaged in 130W-100W
16
STD Biases of Temperature Averaged in 130W-100W
17
RMSE/STD of Upper 300m Heat Content Anomaly
18
Mean Biases of Salinity Averaged in 2S-2N
19
STD Biases of Salinity Averaged in 2S-2N
20
Anomaly Correlation of Surface Zonal Current with OSCAR
21
RMSE/STD of SSH
Summary Coordinated Observing System Experiments (OSEs) were conducted
using NCEP’s GODAS and GFDL’s ECDA during 2004-2011. OSE simulations were evaluated with observations (TAO
temperature, TAO currents, altimetry SSH) and analyzed data (EN4 temperature and salinity analysis, OSCAR surface current analysis)
Without assimilation of in situ data, both GODAS and ECDA had large mean biases, STD biases and RMSE.
Assimilation of in situ data significantly reduced mean biases, STD biases and RMSE in all variables except zonal current at equator.
For constraining temperature analysis, the mooring data is more critical than the Argo data in the equatorial Pacific, but the Argo data is more important in off-equatorial regions.
For constraining salinity, sea surface height and surface current analysis, the influence of Argo data is more critical.
Impacts of observations on ocean analysis are sensitive to configuration of data assimilation schemes, indicating that a multi-model approach is needed to assess the value of tropical Pacific observing systems.
Extend CLIVAR-GSOP/GODAE OceanView Ocean Reanalyses Intercomparison Project (ORA-IP) into real time
Assess uncertainties in temperature analysis of tropical Pacific in support of ENSO monitoring and prediction
Explore any connections between gaps in ocean observations and spreads among ensemble ORAs
Articulate needs for sustained ocean observing systems in support of TPOS2020
Monitor signal-to-noise ratio in the global ocean temperature, 300m heat content, depth of 20C isotherm
23
Real-Time Ocean Renalyses Intercomparison
See details in the poster by Xue et al.
24
Hindcast skill of RMSE difference normalized by RMSE of ALL for CFSv2 (left) and CM2.1 (right) for various SST indices. Red bar: noMoor minus ALL; Green bar: noArgo minus ALL; Blue bar: CTL minus ALL. Filled (unfilled) bars are skill for L0-L4 (L5-L9). The RMSE are calculated with seasonal mean SST anomalies between model and observations for all initial months, all years (2004-2011) and lead month L0-L4 and L5-L9 separately. RMSEs are calculated for NINO3, NINO4 (5oS-5oN, 160oE-150oW), and NINO3.4 (5oS-5oN, 170-120oW) regions, tropical Pacific (TPAC; 20oS-20oN, 120oE-80oW), tropical Indian (TIND; 20oS-20oN, 30-120oE), western tropical Indian (WTIO; 10oS-10oN, 50-70oE), and south eastern tropical Indian (SETIO; 10oS-0o, 90-110oE) Oceans, equatorial Atlantic (EQATL; 5oS-5oN, 70oW-30oE), and Main Developing Region (MDR; 10-20oN, 80-20oW).
25
26
27
28