Search

This work aims at determining the best performing mid and far-infrared (MIR and FIR) joint spectral interval to identify and classify clouds in the Antarctic region by mean of a machine learning algorithm.About 1700 spectral-resolved radiances, collected during 2013 by the ground based Radiation Explorer in the Far InfraRed-Prototype for Applications and Development, REFIR-PAD (Palchetti et al., 2015) at Dome C, Antarctic Plateau, are selected in coincidence with the co-located with backscatter and depolarization profiles derived from a tropospheric lidar system (Ricaud et al., 2017) to pre-classify clear sky, ice clouds, or mixed phase clouds.A machine learning cloud identification and classification algorithm named CIC (Maestri et al., 2019), trained with a pre-selected set of REFIR-PAD spectra, is applied to this dataset by assuming that no other information than the spectrum itself is known.The CIC algorithm is applied by considering different spectral intervals, in order to maximize the classification results for each class (clear sky, ice clouds, mixed phase clouds). A CIC "threat score" is defined as the classification true positives divided by the sum of true positives, false positives, and false negatives. The maximization of the threat score is used to assess the algorithm performances that span from 58% to 96% in accordance with the selected interval. The best performing spectral range is the 380-1000 cm-1. The result, besides suggesting the importance of a proper algorithm calibration in accordance with the used sensor, highlights the fundamental role of the FIR part of the spectrum.The calibrated CIC algorithm is then applied to a larger REFIR-PAD dataset of about 90000 spectra collected from 2012 to 2015. Some results of the full dataset cloud classification are also presented.The present work contributes to the preparatory studies for the Far-infrared Outgoing Radiation Understanding and Monitoring (FORUM) mission that has recently been selected as ESA’s 9th Earth Explorer mission, scheduled for launch in 2026. References:Maestri, T., Cossich, W., and Sbrolli, I., 2019: Cloud identification and classification from high spectral resolution data in the far infrared and mid-infrared, Atmos. Meas. Tech., 12, pp. 3521 - 3540Palchetti, L., Bianchini, G., Di Natale, G., and Del Guasta, M., 2015: Far infrared radiative properties of water vapor and clouds in Antarctica. Bull. Amer. Meteor. Soc., 96, 1505–1518, doi: http://dx.doi.org/10.1175/BAMS-D-13-00286.1.Ricaud, P., Bazile, E., del Guasta, M., Lanconelli, C., Grigioni, P., and Mahjoub, A., 2017: Genesis of diamond dust, ice fog and thick cloud episodes observed and modelled above Dome C, Antarctica, Atmos. Chem. Phys., 17, 5221–5237, https://doi.org/10.5194/acp-17-5221-2017.

Far- to mid- infrared spectral radiances measured either from high altitude platforms or from ground can be processed to retrieve atmospheric vertical profiles and cloud parameters, variables particularly relevant in climate change studies. The retrieval requires a forward model with the capability of simulating the multiple scattering from cloud particles. The Discrete Ordinate Radiative Transfer (DISORT) offers this possibility, however, accurate simulations can be obtained only with a huge computational load. We developed a forward / retrieval model based on the two-streams δ-Eddington approximation, allowing much faster computations, while retaining good accuracy. The code, named SACR (Simultaneous Atmospheric and Clouds Retrieval), allows to retrieve simultaneously temperature and gas profiles, cloud micro-physical and geometrical parameters and surface temperature from vertical sounding observations. We illustrate the equations implemented in the SACR code, prove the self-consistency of the inversion and assess its forward model accuracy with a focus on the range from 200 to 1000 cm − 1 . The assessment is made by comparing the simulated spectral radiances to those computed by LBLDIS, a very accurate model integrating LBLRTM (Line By Line Radiative Transfer Model) and DISORT. For cloud particle sizes between 20 and 100 µm and optical depths between 0.1 and 10, our model shows biases smaller than 0.4 mW/(m2 sr cm − 1 ) in upwelling radiance simulations, and biases smaller than 0.3 mW/(m2 sr cm − 1 ) in downwelling radiance simulations. Depending on the spectral grid and on the number of atmospheric layers used, the SACR code is from 5 to 8 times faster than LBLDIS.

Since December, 2011 an integrated observation system based on the REFIR-PAD Fourier transform spectrora- diometer has been established at the station Concordia, on the East Antarctic plateau (75 o 06'S, 123 o 23'E, 3233 m a. s. l.). The REFIR-PAD instrument provides spectrally-resolved measurements of the downwelling atmospheric emitted radiance in the 100-1500 cm -1 range, thus including the far-infrared region; together with a set of auxiliary sensors connected to a single data acquisition and storage system these measurements provide a valuable tool for the characterization of the Antarctic troposphere. The observation system is operating in continuous acquisition mode since its installation, thus providing an unique, more than 7-year long, time series of atmospheric radiances and ancillary data. Polar regions are the main radiative sink of the Earth system, thus the monitoring of their atmosphere has a high rel- evance for climate studies. Broad band, spectrally resolved radiance measurements as those performed by REFIR- PAD fulfill a twofold task, providing a characterization of the atmospheric radiation including the identification of its main contributing components, but also allowing for the retrieval of the atmospheres structure and composition through the analysis of the measured spectra. This kind of analysis is particularly interesting in the polar context, since it includes the far-infrared region of the spectrum in which, due to the low temperatures involved, a great part of the radiative exchange takes place. A description of the currently operating observation system will be presented, together with some preliminary results from the analysis of the measurements performed in the 2012-2018 period.

Cirrus clouds play a key role in the Earth's radiation budget but there is a large uncertainty about their infrared ra- diative impact due to the high variability in the size/shape distribution of ice particles. Their radiative effect is very strong in the atmospheric window between 820-960 cm -1 (8-12 ?m) but the contribution in the far infrared (FIR) of the spectrum, below 667 cm -1 (> 15 ?m), is also very important. In fact about 50% of the entire thermal flux is emitted in the FIR spectral range and here the sensitivity to the ice particle properties, such as crystal habits, is very large. Despite this, only few measurements exist in the FIR that could be used to improve the characterisation of cirrus radiative effect. The present work describes remote sensing data concerning high latitudes ice clouds and the derivation of their optical and micro-physical properties, such as the crystals effective diameter, the ice water path and optical depth. Measurements are performed by means of a Fourier transform spectroradiometer, named Radiation Explorer Far Infrared - Prototype for Applications and Development (REFIR-PAD), operating in the FIR spectral region. This instrument has been successfully used during many field campaigns and it is currently installed at Concordia base at Dome-C (Antarctica, 3230 m a.s.l.), where has been operating continuously in ground-based zenith-looking ob- servation geometry since 2012. This campaign has allowed to collect a wide database of the downwelling spectral radiance, also in presence of cirrus clouds, together with other supporting measurements, such as lidar and ther- modynamic vertical profiles, for independent characterisation of the atmospheric state. The retrieval of the cirrus clouds parameters has been performed by two independent methodologies: the physical Radiative Transfer - Retrieval (RT-RET) and the Simultaneous Atmospheric and Clouds Retrieval (SACR) code which is based on optimal estimation. This last code, developed for this purpose, allows to perform the retrieval of the atmospheric variables, such as the vertical profiles of water vapour and temperature, and the cirrus clouds properties by using the single scattering coefficients of different ice crystal habits, provided by specific databases. The analysis of the cirrus clouds properties, performed by using the ground-based measurements of REFIR-PAD, will be of support to the Far-infrared-Outgoing-Radiation Understanding and Monitoring (FORUM) mission, re- cently selected by ESA as candidate for the EE-9 missions. FORUM will perform for the first time the spectral measurement of the FIR outgoing radiation and, by flying in tandem with IASI-NG which measures the mid-IR component, will open up a new window for understanding and quantifying the radiative processes involving water vapour, clouds and other greenhouse gases.

The possibility separating the contributions of the atmospheric state and ice clouds by using spectral infrared measurements is a fundamental step to quantifying the cloud effect in climate models. A simultaneous retrieval of cloud and atmospheric parameters from infrared wideband spectra will allow the disentanglement of the spectral interference between these variables. In this paper, we describe the development of a code for the simultaneous retrieval of atmospheric state and ice cloud parameters, and its application to the analysis of the spectral measurements acquired by the Radiation Explorer in the Far Infrared-Prototype for Applications and Development (REFIR-PAD) spectroradiometer, which has been in operation at Concordia Station on the Antarctic Plateau since 2012. The code performs the retrieval with a computational time that is comparable with the instrument acquisition time. Water vapour and temperature profiles and the cloud optical and microphysical properties, such as the generalised effective diameter and the ice water path, are retrieved by exploiting the 230-980?EURcm?'1 spectral band. To simulate atmospheric radiative transfer, the Line-By-Line Radiative Transfer Model (LBLRTM) has been integrated with a specifically developed subroutine based on the Î-Eddington two-stream approximation, whereas the single-scattering properties of cirrus clouds have been derived from a database for hexagonal column habits. In order to detect ice clouds, a backscattering and depolarisation lidar, co-located with REFIR-PAD has been used, allowing us to infer the position and the cloud thickness to be used in the retrieval. A climatology of the vertical profiles of water vapour and temperature has been performed by using the daily radiosounding available at the station at 12:00?EURUTC. The climatology has been used to build an a priori profile correlation to constrain the fitting procedure. An optimal estimation method with the Levenberg-Marquardt approach has been used to perform the retrieval. In most cases, the retrieved humidity and temperature profiles show a good agreement with the radiosoundings, demonstrating that the simultaneous retrieval of the atmospheric state is not biased by the presence of cirrus clouds. Finally, the retrieved cloud parameters allow us to study the relationships between cloud temperature and optical depth and between effective particle diameter and ice water content. These relationships are similar to the statistical correlations measured on the Antarctic coast at Dumont d'Urville and in the Arctic region.