Tier 2 Clausius–Clapeyron Test

1 figure · 1990–2014 → 2040–2049

Synthesis

Under SSP3-7.0, coupled IFS configurations exhibit a global water vapor scaling of 4.1–4.5%/K, deviating from the theoretical 7%/K Clausius-Clapeyron rate due to marked relative humidity reductions in continental and high-warming regimes.
Analysis of the Clausius-Clapeyron scaling in IFS-FESOM and IFS-NEMO simulations under SSP3-7.0 reveals a global hydrological sensitivity of 4.1%/K and 4.5%/K, respectively. This global sub-Clausius-Clapeyron response significantly deviates from the theoretical ~7%/K expectation derived under the assumption of constant relative humidity. The divergence is driven by a distinct bimodal regime in the thermodynamic response: while oceanic regions largely adhere to the 7%/K scaling implied by unlimited moisture availability, continental regions exhibit a suppressed moisture response, dragging down the global aggregate. The scatter distribution highlights a decoupling of moisture capacity and availability in regions experiencing high warming magnitudes (ΔT > 4 K). In these regimes, the fractional increase in Total Column Water Vapor (TCWV) saturates, indicating substantial reductions in relative humidity. This 'tail' of high-warming, low-moistening grid cells likely corresponds to land surfaces and high-latitude amplification zones where temperature rise outpaces advective moisture supply. While both coupled configurations share this structural behavior, IFS-NEMO exhibits a slightly stronger hydrological sensitivity, suggesting subtle differences in surface flux parameterizations or land-sea warming ratios compared to IFS-FESOM.

Related diagnostics

hurs_spatial_map tas_spatial_map pr_global_tseries

Clausius–Clapeyron Test

Clausius–Clapeyron Test
Variables avg_tcwv, avg_2t
Models ifs-fesom, ifs-nemo
Units kg/m2
Baseline 1990-2014
Future 2040-2049
Method δ_TCWV = ln(TCWV_future / TCWV_hist); ΔT = T_future − T_hist. Sub-sampled to 50k cells.

Summary high

This diagnostic evaluates the hydrological sensitivity of two coupled models (IFS-FESOM and IFS-NEMO) by correlating fractional changes in Total Column Water Vapor (TCWV) with surface temperature warming (ΔT) between the 1990s and 2040s.

Key Findings

  • Both models exhibit a global moisture scaling rate (4.1%/K for IFS-FESOM, 4.5%/K for IFS-NEMO) that is significantly lower than the theoretical Clausius-Clapeyron (C-C) rate of 7%/K.
  • A distinct bifurcation occurs in the data: a primary cluster follows the C-C prediction (likely oceanic regions maintaining constant relative humidity), while a secondary distribution shows suppressed moisture increase (likely land regions).
  • At high warming magnitudes (ΔT > 4 K), the moisture response saturates, deviating strongly from the linear 7%/K expectation, indicating a decoupling of moisture capacity and availability in rapidly warming regions.

Spatial Patterns

The scatter plot structure implies two distinct physical regimes: moisture-unlimited regions (oceans) where data points align closely with the red 7%/K C-C line, and moisture-limited regions (continents/arid zones) where the scaling is shallower. The 'tail' of points extending to high ΔT (up to 6-7 K) with low fractional TCWV increase is characteristic of high-latitude amplification or continental interiors where warming outpaces moisture advection.

Model Agreement

The models are qualitatively similar in distribution structure, confirming consistent thermodynamic responses. However, IFS-NEMO shows a slightly stronger hydrological sensitivity (4.5%/K vs 4.1%/K), suggesting differences in surface latent heat flux parameterizations or a slightly different land-sea warming ratio compared to IFS-FESOM.

Physical Interpretation

The theoretical C-C rate of ~7%/K assumes constant Relative Humidity (RH). The observed global sub-C-C scaling is driven by deviations from constant RH, particularly over land (where soil moisture limits evaporation) and in regions of strong warming amplification where specific humidity increases cannot match the pace of temperature rise. The flattening at high ΔT physically represents a reduction in RH in the most rapidly warming grid cells.

Caveats

  • The single linear fit oversimplifies the bimodal nature of the response (land vs. ocean regimes).
  • Sub-sampling to 50k cells may under-represent extreme events or small-scale features relevant to high-resolution modeling.