A Feedforward Neural Network approach for the detection of optically thin cirrus from IASI-NG

The identification of optically thin cirrus is crucial for their accurate parameterization in climate and Earth’s system models. This study exploits the characteristics Infrared Atmospheric Sounding Interferometer -New Generation (IASI-NG) to develop an algorithm detection cirrus. IASI-NG has been designed EUMeTsAT Polar System Second program continue service its predecessor IASI from 2024 onward. A Thin Cirrus Detection Algorithm (TCDA) presented here, as developed IASI-NG, but also parallel evaluate performance on currently-available real observations. TCDA uses a Feedforward Neural Network (NN) approach detect eventually misidentified clear sky by previously applied cloud algorithm. estimates uncertainty “clear sky” or “thin cirrus” detection. NN trained tested dataset (or IASI) simulations obtained processing ECMWF 5-generation reanalysis (ERA5) data with σ-IASI radiative transfer model. validation against independent simulated provides quantitative statistical assessment improvements brought respect IASI. In fact, outperforms 3% POD, 1% bias, 2% accuracy FAR passes 0.02 0.01. Moreover, state-of-the-art products Cloudsat/CPR CALIPSO/CALIOP observations, reveals tendency underestimate presence (POD=0.47) low (0.07) which drops 0.0 very