Evaluating the ECHAM5/MPI-OM Coupled Climate Model: Wind Stress Parameterization, Mean Ocean Circulation, and Tropical Variability
Observation and Context Coupled atmosphere-ocean general circulation models often struggle to simulate tropical climate accurately. Previous iterations developed by the Max Planck Institute for Meteorology (MPI-M) and other climate modeling centers produced a severe equatorial “cold bias,” where simulated Pacific sea surface temperatures were much colder than observed. Standard models usually calculate surface wind stress assuming the ocean surface is motionless, even though equatorial surface currents can move rapidly (exceeding 1 m/s) in the same direction as trade winds.
Hypothesis Scientists hypothesized that accounting for the relative velocity between wind and ocean surface currents in wind stress calculations would weaken trade wind stress, slow equatorial ocean divergence and upwelling, and substantially reduce the equatorial Pacific cold bias while improving the simulation of El Niño–Southern Oscillation (ENSO) variability without requiring artificial flux adjustments.
Experiment and Methodology The researchers constructed and tested the coupled ECHAM5/MPI-OM model:
- Model Configuration: The atmospheric component (ECHAM5) ran at T63 spectral resolution (approximately 1.875∘×1.875∘) with 31 vertical levels. The ocean model (MPI-OM) featured an average 1.5∘ horizontal resolution with 40 vertical levels on an orthogonal curvilinear grid, placing the grid poles over Greenland and Antarctica to resolve deep-water formation regions without polar coordinate singularities.
- Testing Conditions: Scientists ran two 300-year present-day control experiments. The primary experiment included the Wind Stress Correction (WSC), calculating stress from the velocity difference between near-surface winds and ocean currents. This was compared directly to a parallel run with No Wind Stress Correction (NWSC). Model years 250–300 were evaluated against observational datasets.
Results and Data
- Equatorial Cold Bias: Implementing WSC reduced equatorial westward wind stress, causing an estimated 30% reduction in equatorial upwelling (lowering it to approximately 2×10−5 m/s at a depth of 50 m). This warmed the central and western equatorial Pacific by more than 1 K, drastically reducing the model’s cold bias.
- Tropical Precipitation: Warmer western Pacific waters intensified atmospheric convection, diminishing the equatorial dry bias by up to 2.5 mm/day (a roughly 30% reduction).
- ENSO Characteristics: Under WSC, sea surface temperature variability in the eastern and central equatorial Pacific weakened by about 30%, suppressing excessive westward anomaly propagation and shifting ENSO toward a realistic standing thermocline mode. The dominant ENSO periodicity shifted from an unrealistic 3 years to 4 years, aligning closely with real-world observations.
- Large-Scale Ocean Performance: The model produced a stable North Atlantic Deep Water overturning circulation of 18.5 Sv (matching the observed 18±4 Sv) and realistic Arctic sea ice exports through Fram Strait (2,618 km3/year).
- Remaining Deficiencies: WSC did not resolve warm sea surface temperature biases along the western coasts of South America and southern Africa, and an erroneous double Intertropical Convergence Zone (ITCZ) remained.
Conclusion and Climate Impact Accounting for moving ocean currents in surface wind stress calculations significantly improves simulated tropical climate dynamics. This parameterization corrects longstanding equatorial ocean upwelling errors, relieves the cold sea surface temperature bias, and generates realistic ENSO cycles without resorting to artificial flux adjustments.
- Climate Impact: By producing a well-balanced global heat and freshwater transport system and capturing authentic multi-year tropical climate variability, ECHAM5/MPI-OM served as a core prototype model for multicentury climate change projections evaluated in the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4).
Full Citation Jungclaus, J. H., Keenlyside, N., Botzet, M., Haak, H., Luo, J.-J., Latif, M., Marotzke, J., Mikolajewicz, U., & Roeckner, E. (2006). Ocean circulation and tropical variability in the coupled model ECHAM5/MPI-OM. Journal of Climate, 19(16), 3952–3972. https://doi.org/10.1175/JCLI3827.1