Skip to content

Author

R. Roehrig

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Jul 2026

Evaluation of the Infrared Spectral Signature of ARPEGE‐Climat 6.3 in Clear‐Sky Conditions Using IASI Observations

The Earth's top‐of‐atmosphere (TOA) radiative budget is a key feature of the Earth's climate, and broadband TOA observations are widely used to evaluate climate models. However, recent studies have demonstrated that a good match between simulated and observed broadband fluxes can result from error compensation in the spectral dimension. Hence, assessing the spectrally resolved radiative budget is promising. Here, we use 7 years of IASI hyperspectral infrared observations to evaluate ARPEGE‐Climat, the atmospheric component of the CNRM‐CM6 climate model. To this end, synthetic nadir‐looking IASI spectra are computed from the atmospheric profiles and surface properties of an ARPEGE‐Climat amip simulation, using the radiative transfer code RTTOV, and compared to IASI observations for clear‐sky conditions. Over the ocean, ARPEGE‐Climat exhibits a cold brightness temperature bias (∼1 K) in the atmospheric window, attributed to inappropriate sea surface emissivity and sea ice temperature. A cold bias is also found in the water vapor absorption band due to a wet and cold bias in the troposphere. The cold bias is stronger in the C O2 ${\mathrm{O}}_{2}$ band because the cold bias is more pronounced in the stratosphere. Over land, the cold biases are stronger (∼5 K in the atmospheric window), and mostly due to issues in the surface emissivity and skin temperature at high latitudes and above elevated terrain. The mean annual cycle in selected IASI channels is well captured by ARPEGE‐Climat, although slightly underestimated in amplitude. This study demonstrates the potential of hyperspectral infrared observations for objective climate model evaluation.

L. Leonarski, R. Roehrig, S. Della Fera et al. · 0 citations