Tier 3 Global Mean Time Series CMIP6

3 figures · 1990–2014 → 2040–2049

CMIP6 Multi-Model Mean Context

Comparison with CMIP6 conventional-resolution ensemble mean (up to 8 models under SSP3-7.0, regridded to 0.25°).

Contributing models: MPI-ESM1-2-LR/r1i1p1f1, MPI-ESM1-2-LR/r2i1p1f1, MPI-ESM1-2-LR/r3i1p1f1, MPI-ESM1-2-LR/r4i1p1f1, MPI-ESM1-2-LR/r5i1p1f1, GISS-E2-1-G/r1i1p1f2, GISS-E2-1-G/r2i1p1f2, GISS-E2-1-G/r3i1p1f2, GISS-E2-1-G/r4i1p1f2, GISS-E2-1-G/r5i1p1f2, IPSL-CM6A-LR/r1i1p1f1, IPSL-CM6A-LR/r2i1p1f1, IPSL-CM6A-LR/r3i1p1f1, IPSL-CM6A-LR/r4i1p1f1, IPSL-CM6A-LR/r5i1p1f1, ACCESS-ESM1-5/r1i1p1f1, ACCESS-ESM1-5/r2i1p1f1, ACCESS-ESM1-5/r3i1p1f1, ACCESS-ESM1-5/r4i1p1f1, ACCESS-ESM1-5/r5i1p1f1, EC-Earth3/r1i1p1f1, EC-Earth3/r4i1p1f1, EC-Earth3/r6i1p1f1, EC-Earth3/r9i1p1f1, CNRM-CM6-1/r1i1p1f2, CNRM-CM6-1/r2i1p1f2, CNRM-CM6-1/r3i1p1f2, CNRM-CM6-1/r4i1p1f2, CNRM-CM6-1/r5i1p1f2, AWI-CM-1-1-MR/r1i1p1f1, AWI-CM-1-1-MR/r2i1p1f1, AWI-CM-1-1-MR/r3i1p1f1, AWI-CM-1-1-MR/r4i1p1f1, AWI-CM-1-1-MR/r5i1p1f1, CNRM-ESM2-1/r1i1p1f2, CNRM-ESM2-1/r2i1p1f2, CNRM-ESM2-1/r3i1p1f2, CNRM-ESM2-1/r4i1p1f2, CNRM-ESM2-1/r5i1p1f2, FGOALS-g3/r1i1p1f1, FGOALS-g3/r2i1p1f1, FGOALS-g3/r3i1p1f1, FGOALS-g3/r4i1p1f1, FGOALS-g3/r5i1p1f1, INM-CM5-0/r1i1p1f1, INM-CM5-0/r2i1p1f1, INM-CM5-0/r3i1p1f1, INM-CM5-0/r4i1p1f1, INM-CM5-0/r5i1p1f1, MRI-ESM2-0/r1i1p1f1, MRI-ESM2-0/r2i1p1f1, MRI-ESM2-0/r3i1p1f1, MRI-ESM2-0/r4i1p1f1, MRI-ESM2-0/r5i1p1f1

Synthesis

DestinE models demonstrate global warming trends, hydrological sensitivity, and radiative forcing responses that are statistically indistinguishable from the CMIP6 ensemble median, validating their suitability for projection under SSP3-7.0.
The IFS-FESOM and IFS-NEMO simulations exhibit robust stability and physically consistent transient responses to SSP3-7.0 forcing across all global mean metrics. For near-surface temperature, both models remain strictly within the CMIP6 P5–P95 uncertainty envelope; however, a systematic offset persists where IFS-FESOM tracks near the ensemble median, while IFS-NEMO occupies the lower quartile (approximately 0.5–0.8 K cooler). Despite this difference in absolute mean state—likely attributable to initialization and ocean core formulations—the warming rates in both high-resolution configurations are nearly identical to the CMIP6 median, confirming that the effective transient climate sensitivity is standard. Hydrological sensitivity is similarly well-captured, with global precipitation scaling with warming at rates parallel to the CMIP6 median. Both models display a minor wet bias (~1–2%) relative to the multi-model mean, likely reflecting enhanced resolution of convective processes, yet they remain well below the P95 bound ($f_{out} \approx 0\%$). The Top-of-Atmosphere (TOA) radiative imbalance provides definitive validation of the coupled system's energy budget, tracking the CMIP6 median closely with a persistent positive imbalance of 1.0–2.0 W/m² indicative of SSP3-7.0 greenhouse trapping, while correctly resolving the transient negative forcing anomaly associated with the 1991 Mt. Pinatubo eruption.

Related diagnostics

tas_bias_map pr_bias_map tos_timeseries

Global Mean 2m Temperature Time Series

Global Mean 2m Temperature Time Series
Variables avg_2t
Models ifs-fesom, ifs-nemo, CMIP6-MMM
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

The time series shows Global Mean 2m Temperature (1990–2050) for IFS-FESOM and IFS-NEMO against the CMIP6 SSP3-7.0 envelope; both models capture the warming trend and fall well within the CMIP6 uncertainty range.

Key Findings

  • IFS-FESOM is systematically warmer than IFS-NEMO by approximately 0.5–0.8 K throughout the simulation period.
  • IFS-FESOM tracks slightly above the CMIP6 median (P50), while IFS-NEMO tracks consistently below the median, occupying the lower half of the CMIP6 envelope.
  • Both DestinE models exhibit a warming rate consistent with the CMIP6 ensemble median under SSP3-7.0, validating their transient climate response.
  • Interannual variability in both high-resolution models is realistic and comparable in magnitude to the CMIP6 range.

Spatial Patterns

N/A (Global mean time series; spatial patterns are not resolved in this figure).

Model Agreement

Both IFS-FESOM and IFS-NEMO show excellent agreement with the CMIP6 ensemble regarding the rate of warming (trend). However, there is a persistent offset in the absolute mean state, with IFS-FESOM initiating and remaining warmer than IFS-NEMO. Both remain strictly within the P5–P95 CMIP6 envelope.

Physical Interpretation

The persistent temperature offset between the two DestinE models is likely driven by differences in the ocean mean state (SSTs, sea ice extent) resulting from the different ocean cores (unstructured FESOM vs. structured NEMO) and their initialization/coupling. The shared upward trend confirms that both coupled systems are responding correctly to the increasing radiative forcing of the SSP3-7.0 scenario.

Caveats

  • The vertical line labeled 'Scenario start' appears at 2040, likely marking the start of the specific 'Future period' defined in metadata rather than the standard CMIP6 scenario branch time (2015).
  • The DestinE simulations cover a shorter period (~1990–2050) than the full CMIP6 range shown.

Global Mean Precipitation Rate Time Series

Global Mean Precipitation Rate Time Series
Variables avg_tprate
Models ifs-fesom, ifs-nemo, CMIP6-MMM
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 compared against the CMIP6 SSP3-7.0 ensemble envelope (1850–2100).

Key Findings

  • Both IFS-FESOM and IFS-NEMO fall well within the CMIP6 P5–P95 envelope, indicating their global mean precipitation climatology is consistent with the broader multi-model ensemble.
  • The DestinE models track slightly above the CMIP6 median (P50), showing a wet bias of approximately 1-2% relative to the median but remaining far from the upper P95 bound.
  • Both high-resolution models exhibit a steady increasing trend in global precipitation from ~2000 onwards, paralleling the gradient of the CMIP6 median response to warming.

Spatial Patterns

Not applicable (figure presents a 1-dimensional global mean time series rather than spatial fields).

Model Agreement

High agreement with CMIP6 expectations. The DestinE models are centrally located within the CMIP6 distribution, suggesting no outliers in global hydrological sensitivity or mean state. IFS-FESOM and IFS-NEMO track each other very closely after an initial adjustment period.

Physical Interpretation

The positive trend in precipitation is a robust response to global warming, driven by the Clausius-Clapeyron relation where specific humidity increases with temperature, intensifying the hydrological cycle. The slightly higher mean precipitation in DestinE compared to the CMIP6 median may reflect improved resolution of precipitation-generating processes (e.g., orographic uplift, resolved convection) that are parameterized in coarser models.

Caveats

  • IFS-NEMO shows a visible initialization shock or adjustment period around 1990 before stabilizing.
  • The analysis window for DestinE (ending ~2050) is shorter than the full CMIP6 projection.

TOA Radiative Imbalance Time Series

TOA Radiative Imbalance Time Series
Variables avg_tnswrf, avg_tnlwrf
Models ifs-fesom, ifs-nemo, CMIP6-MMM
Units W/m2
Baseline 1990-2014
Future 2040-2049
Method DestinE: avg_tnswrf + avg_tnlwrf (net fluxes). CMIP6: rsdt - rsut - rlut (incoming - outgoing). Global mean at each timestep.

Summary high

This time series analysis compares the global mean Top-of-Atmosphere (TOA) radiative imbalance of two DestinE models (IFS-FESOM and IFS-NEMO) against the CMIP6 historical and SSP3-7.0 ensemble envelope from 1990 to 2050.

Key Findings

  • Both IFS-FESOM and IFS-NEMO show excellent agreement with the CMIP6 ensemble median (P50), tracking the multi-model mean closely throughout the simulation period.
  • The models exhibit a persistent positive radiative imbalance of approximately 1.0–2.0 W/m² between 2020 and 2050, consistent with the planetary heating expected under the SSP3-7.0 scenario.
  • A distinct negative excursion is visible at the beginning of the DestinE simulations (~1991), aligning with the radiative forcing reduction caused by the Mt. Pinatubo eruption.
  • High-frequency seasonal oscillations (amplitude ~10 W/m²) are visible in the raw DestinE data, driven by the eccentricity of Earth's orbit (varying solar insolation), while the smoothed trend represents the climate response.

Spatial Patterns

Not applicable (global mean time series). The temporal pattern shows a recovery from the 1991 volcanic dip followed by a steady, slightly increasing positive energy imbalance indicative of continuous greenhouse gas forcing.

Model Agreement

There is high agreement between the two DestinE models (which overlap almost perfectly) and the CMIP6 median. Both DestinE simulations fall well within the CMIP6 P5–P95 envelope, suggesting no anomalous global energy drift or outlier behaviour.

Physical Interpretation

The positive TOA imbalance confirms the models are accumulating heat due to the enhanced greenhouse effect (reduced outgoing longwave radiation relative to absorbed shortwave). The 1991 dip demonstrates the models' response to volcanic stratospheric aerosols (increased reflection of shortwave radiation).

Caveats

  • The simulation start date (1990) coincides with the Pinatubo eruption, making it difficult to completely disentangle potential initialization shock from the physical volcanic response, although the recovery tracks CMIP6 well.
  • The faint vertical lines represent raw monthly data showing the orbital seasonal cycle, which dominates the visual variability compared to the interannual trend.