Tier 3 Global Mean Time Series

3 figures · 1990–2014 → 2040–2049

Synthesis

While both models project robust warming and hydrological intensification under SSP3-7.0, IFS-FESOM is systematically warmer and wetter than IFS-NEMO, which conversely maintains a higher rate of planetary energy accumulation.
The global mean diagnostics for IFS-FESOM and IFS-NEMO under SSP3-7.0 demonstrate a coherent, highly synchronized response to anthropogenic forcing from 1990 to 2050, driven by the shared IFS atmospheric component. Both models exhibit robust warming trends, with global mean 2-meter temperatures rising from a ~287.5 K baseline in the 1990s to over 289.0 K by 2050, accompanied by a consistent intensification of the hydrological cycle and a shift toward positive Top-of-Atmosphere (TOA) radiative imbalance (+1.0 to +1.8 W/m² by the 2040s). The distinct radiative signature of the 1991 Mt. Pinatubo eruption is resolved identically in both simulations, confirming the models' sensitivity to historical volcanic aerosol forcing. Despite similar transient climate sensitivities, a systematic thermodynamic offset persists between the two ocean couplings. IFS-FESOM maintains a warm bias (+0.5–0.8 K) and a wetter mean state (~1.5% higher precipitation) relative to IFS-NEMO throughout the simulation. Physically, these mean-state differences are energetically consistent: the warmer surface in IFS-FESOM facilitates greater outgoing longwave radiation, resulting in a lower net TOA imbalance. Conversely, IFS-NEMO remains cooler at the surface but exhibits a significantly higher net energy uptake (offset by ~0.3–0.5 W/m²), suggesting either greater ocean heat uptake efficiency below the mixed layer or stronger low-cloud cooling feedbacks associated with the structured NEMO grid compared to the unstructured FESOM discretization.

Related diagnostics

global_ocean_heat_content_profiles sea_ice_extent_timeseries surface_air_temperature_spatial_bias

Global Mean 2m Temperature Time Series

Global Mean 2m Temperature Time Series
Variables avg_2t
Models ifs-fesom, ifs-nemo
Units K
Baseline 1990-2014
Future 2040-2049
Method Global mean = spatial .mean() (HealPix equal-area). Historical and scenario concatenated. Trend fit over scenario period only.

Summary high

Time series of global mean 2m temperature (1990–2050) comparing IFS-FESOM and IFS-NEMO, revealing a persistent warm offset in the FESOM coupling despite similar long-term warming trends under SSP3-7.0.

Key Findings

  • IFS-FESOM exhibits a systematic warm bias of approximately 0.5–0.8 K relative to IFS-NEMO throughout the entire simulation period.
  • Both models show a robust warming trend consistent with anthropogenic forcing, rising from a baseline of ~287.5 K in the 1990s to ~289.0–289.5 K by 2050.
  • Interannual variability is pronounced in both models, with decadal fluctuations occasionally diverging (e.g., 2025–2035) before converging on positive trends in the 2040–2050 analysis window.

Spatial Patterns

While spatially aggregated, the persistent global offset implies widespread differences in Sea Surface Temperature (SST) or sea-ice extent distributions between the unstructured FESOM grid and the structured NEMO grid, likely affecting global albedo.

Model Agreement

Models agree on the seasonal cycle amplitude and the direction of the climate change signal (warming). Disagreement lies in the absolute mean state (IFS-FESOM is consistently warmer) and the phasing of internal decadal variability.

Physical Interpretation

The long-term upward trend is driven by radiative forcing from SSP3-7.0 greenhouse gas emissions. The temperature offset between models is likely due to differences in ocean heat uptake efficiencies or sea-ice albedo feedbacks inherent to the different ocean dynamical cores (finite-element FESOM vs. finite-difference NEMO).

Caveats

  • The 'Scenario start' marker at 2040 likely denotes the window for the specific trend analysis shown, rather than the actual SSP forcing start (typically 2015).
  • Fitting a linear trend over a short 10-year window (2040–2050) makes the slope highly sensitive to phases of internal variability (e.g., ENSO) rather than solely reflecting the forced response.
  • ICON data mentioned in the prompt context is absent from this specific figure.

Global Mean Precipitation Rate Time Series

Global Mean Precipitation Rate Time Series
Variables avg_tprate
Models ifs-fesom, ifs-nemo
Units kg/m2/s
Baseline 1990-2014
Future 2040-2049
Method Global mean = spatial .mean() (HealPix equal-area). Historical and scenario concatenated. Trend fit over scenario period only.

Summary high

Time series analysis of global mean precipitation rates (1990–2050) for IFS-FESOM and IFS-NEMO coupled models, illustrating interannual variability and emerging positive trends under the SSP3-7.0 scenario.

Key Findings

  • IFS-FESOM exhibits a systematically higher global mean precipitation rate (mean approx. 3.48e-5 kg m⁻² s⁻¹) compared to IFS-NEMO (mean approx. 3.44e-5 kg m⁻² s⁻¹) throughout the simulation period.
  • Both models show a robust positive trend in precipitation intensity starting circa 2020 and continuing through 2050, consistent with expected hydrological cycle intensification.
  • Decadal variability is notably synchronized between the two models (e.g., concurrent minima around 1994 and 2012), despite the offset in absolute magnitude.

Spatial Patterns

Not applicable; the figure presents a globally averaged scalar time series, masking regional spatial heterogeneity.

Model Agreement

The models show high agreement on the phase of interannual-to-decadal variability and the sign of the long-term trend, likely driven by the shared IFS atmospheric component and external forcing. They disagree on the mean state, with IFS-FESOM consistently wetter by approximately 1.5%.

Physical Interpretation

The upward trend is driven by the energetic constraints of the global hydrological cycle under warming (precipitation increasing at ~2-3% per K to balance radiative cooling). The systematic offset suggests differences in mean Sea Surface Temperature (SST) or global evaporation rates resulting from the different ocean discretizations (FESOM vs. NEMO) interacting with the atmosphere.

Caveats

  • The 'Scenario start' marker at 2040 is non-standard (typically 2015 for SSPs), indicating a specific analysis window rather than the actual divergence of historical/scenario forcing.
  • The ICON model mentioned in the system description is absent from this specific visual.

TOA Radiative Imbalance Time Series

TOA Radiative Imbalance Time Series
Variables avg_tnswrf, avg_tnlwrf
Models ifs-fesom, ifs-nemo
Units W/m2
Baseline 1990-2014
Future 2040-2049
Method TOA imbalance = avg_tnswrf + avg_tnlwrf. Global mean at each timestep.

Summary high

Time series of global-mean Top-of-Atmosphere (TOA) radiative imbalance (1990–2050) for IFS-FESOM and IFS-NEMO, showing the system's energy response to historical volcanic forcing and future SSP3-7.0 greenhouse gas forcing.

Key Findings

  • A distinct negative radiative imbalance excursion (~-2.0 to -2.5 W/m²) occurs in the early 1990s, corresponding to the Mt. Pinatubo eruption.
  • Both models exhibit a positive trend in energy imbalance, rising from near-neutral in the late 1990s to approximately +1.0 to +1.8 W/m² by the 2040s.
  • IFS-NEMO consistently maintains a higher radiative imbalance (offset by ~0.3–0.5 W/m²) compared to IFS-FESOM throughout the entire simulation period.

Spatial Patterns

Not applicable (Global Mean Time Series). The data represents spatially aggregated global energy budget evolution rather than regional variability.

Model Agreement

The models show high temporal correlation in interannual variability (e.g., volcanic response, ENSO-like fluctuations). However, there is a systematic bias where IFS-NEMO simulates a larger net energy uptake than IFS-FESOM. This persistent offset likely stems from differences in ocean heat uptake efficiency or mean state sea surface temperatures (SSTs) between the FESOM (unstructured) and NEMO (structured) ocean components.

Physical Interpretation

The long-term positive trend indicates planetary energy accumulation driven by increasing anthropogenic radiative forcing (SSP3-7.0). The initial dip reflects short-term negative forcing from volcanic stratospheric aerosols (Pinatubo). The persistent positive imbalance in the 2040s confirms the system is not in equilibrium and is continuing to warm. The difference between models implies that IFS-NEMO is either reflecting less shortwave radiation or, more likely, radiating less longwave energy (potentially due to cooler effective surface temperatures or different cloud feedbacks) compared to IFS-FESOM.

Caveats

  • The 'Scenario start' vertical line is positioned at 2040, which is non-standard for CMIP6 (usually 2015), possibly indicating the start of a specific analysis window rather than the forcing scenario change.
  • A sharp drop at the very end of the time series (2050) is likely a filtering/smoothing artifact.
  • Only IFS-FESOM and IFS-NEMO are displayed; the ICON model mentioned in the experimental context is absent from this specific figure.