ALBERT

Description: Cloud top height retrieval from imager instruments is important for nowcasting and for satellite climate data records. A neural network approach for cloud top height retrieval from the imager instrument MODIS (Moderate Resolution Imaging Spectroradiometer) is presented. The neural networks are trained using cloud top layer pressure data from the CALIOP (Cloud-Aerosol Lidar with Orthogonal Polarization) dataset. Results are compared with two operational reference algorithms for cloud top height: the MODIS Collection 6 Level 2 height product and the cloud top temperature and height algorithm in the 2014 version of the NWC SAF (EUMETSAT (European Organization for the Exploitation of Meteorological Satellites) Satellite Application Facility on Support to Nowcasting and Very Short Range Forecasting) PPS (Polar Platform System). All three techniques are evaluated using both CALIOP and CPR (Cloud Profiling Radar for CloudSat (CLOUD SATellite)) height. Instruments like AVHRR (Advanced Very High Resolution Radiometer) and VIIRS (Visible Infrared Imaging Radiometer Suite) contain fewer channels useful for cloud top height retrievals than MODIS, therefore several different neural networks are investigated to test how infrared channel selection influences retrieval performance. Also a network with only channels available for the AVHRR1 instrument is trained and evaluated. To examine the contribution of different variables, networks with fewer variables are trained. It is shown that variables containing imager information for neighboring pixels are very important. The error distributions of the involved cloud top height algorithms are found to be non-Gaussian. Different descriptive statistic measures are presented and it is exemplified that bias and SD (standard deviation) can be misleading for non-Gaussian distributions. The median and mode are found to better describe the tendency of the error distributions and IQR (interquartile range) and MAE (mean absolute error) are found to give the most useful information of the spread of the errors. For all descriptive statistics presented MAE, IQR, RMSE (root mean square error), SD, mode, median, bias and percentage of absolute errors above 0.25, 0.5, 1 and 2 km the neural network perform better than the reference algorithms both validated with CALIOP and CPR (CloudSat). The neural networks using the brightness temperatures at 11 and 12 µm show at least 32 % (or 623 m) lower MAE compared to the two operational reference algorithms when validating with CALIOP height. Validation with CPR (CloudSat) height gives at least 25 % (or 430 m) reduction of MAE.

Description: Cloud top height retrieval from imager instruments is important for Nowcasting and for satellite climate data records. A neural network approach for cloud top height retrieval from the imager instrument MODIS is presented. The neural networks are trained using cloud top layer pressure data from the CALIOP dataset. Results are compared with two operational reference algorithms for cloud top height: the MODIS Collection 6 level 2 height product and the cloud top temperature and height algorithm (CTTH) in the 2014 version of the NWCSAF Polar Platform System (PPS-v2014). All three techniques are evaluated using both CALIOP and CPR (CloudSat) height. Instruments like AVHRR and VIIRS contain fewer channels useful for cloud top height retrievals than MODIS, therefore several different neural networks are investigated to test how infrared channel selection influences retrieval performance. Also a network with only channels available for the AVHRR1 instrument is trained and evaluated. To examine the contribution of different variables, networks with fewer variables are trained. It is shown that variables containing imager information for neighbouring pixels are very important. Overall results for the neural network height retrievals are very promising. The neural networks using the brightness temperatures at 11 μm and 12 μm show at least 33 % (or 627 m) lower mean absolute error (MAE) compared to the two operational reference algorithms when validating with CALIOP height. Validation with CPR (CloudSat) height gives at least 25 % (or 433 m) reduction of MAE. For the network trained with a channel combination available for AVHRR1, the MAE is at least 542 m better when validated with CALIOP and 414 m when validated with CPR (CloudSat) compared to the two operational reference algorithms. The NWCSAF PPS-2018 release will contain a neural network based cloud height algorithm.

Description: Simplified representations of spatially inhomogeneous (three-dimensional (3-D)) clouds in radiative transfer models provide systematic errors when calculating solar broadband radiative fluxes. An example is the neglect of horizontal photon transports as it is the case for the independent column approximation (ICA). The present work tries to quantify and interpret these errors on the basis of a large set of 3-D mixed phase cloud scenarios with 3-D varying extinction coefficients, scattering phase functions, and single-scattering albedos. The cloud cases result from a mesoscale atmospheric circulation model with detailed cloud microphysics. Domain-averaged cloud radiative fluxes are calculated by means of a Monte Carlo radiative transfer model. Depending on cloud type and solar zenith angle (SZA) the differences between 3-D and ICA results range from +20 W m−2 to −30 W m−2 for the upward reflected fluxes and from +10 W m−2 to −7 W m−2 for the absorbed fluxes. The mean (averaged over all cloud realizations) errors of the ICA-based upward fluxes vary between 5 W m−2 overestimation at 15°SZA and 6 W m−2 underestimation at 75°SZA. The ICA underestimates the absorbed flux by ∼1–2 W m−2 for most SZA except for 75°. It is found that neglecting the horizontal variability of the absorption and scattering properties of the cloud hydrometeors leads to a general underestimation of solar broadband absorption by as much as 15 W m−2 with average values between 4 W m−2 at small SZA and 1 W m−2 at large SZA.

Description: The influence of spatially inhomogeneous water vapour distributions on the solar radiative fluxes of inhomogeneous clouds has been estimated by means of Monte Carlo radiative transfer calculations and line by line calculations for transmission at water vapour and oxygen. 96 cloud realizations obtained by the atmospheric model GESIMA have been considered for the following cases: 1) Clouds embedded in a 3d inhomogeneous gas atmosphere and 2) clouds embedded in a stratified gas atmosphere. The resulting differences in solar broadband radiative fluxes appear to be neglectable. Thus, a stratified gaseous atmosphere seems sufficiently adequate for solar radiative transfer calculations. The reasons for these small differences are basically due to the strong correlation between cloud optical thickness and water vapour density as produced by the cloud model used in this study.

Description: In order to investigate the accuracy of simplification in modeling the radiative transfer in those solar spectral regions with major impacts on bio-organisms, i.e., the UVA (0.32–0.4 μm), the UVB (0.28–0.32 μm), and the photosynthetically active radiation (PAR, 0.4–0.7 μm), radiative transfer calculations with varying treatments of cloud geometries (plane-parallel homogeneous (PPHOM), independent column approximation (ICA), and three-dimensional (3-D) inhomogeneous) have been performed. The complete sets of atmospheric information for 133 cloud realizations are taken from the three-dimensional nonhydrostatic mesoscale atmospheric model (GESIMA). A Monte Carlo radiative transfer model (GRIMALDI) has been developed that simulates scattering and absorption for arbitrarily three-dimensional distributions of cloud hydrometeors, air molecules, and water vapor. Results are shown for domain-averaged direct and total transmission (and so, implicitly, diffuse transmission) at the ground surface. In the UVA the PPHOM assumption leads to an underestimation in direct (total) downward flux by as much as 43 (28) W m−2, which is about 49% (32%) of the incoming irradiation, whereas results based on the ICA are almost identical to the 3-D case, except for convective clouds where the error in the UVA for direct (total) downward flux reaches 5 (2) W m−2, or 6% (2%) of the incoming solar irradiation.

Description: The present work investigates to what extend the available cloud informations from large scale, i.e., noncloud-resolving atmospheric models can be used to parameterize the radiative fluxes of 3D clouds. To this end, domain averaged cloud properties from mesocale 3D cloud models arc correlated with domain averaged results from solar broadband 3D radiative transfer calculations for the same clouds. An EOF-analysis shows that albedo, total transmission and absorption (diffuse and direct transmission) are strongest correlated with liquid water path and cloud bottom height (rain water path and cloud cover). A multidimensional quasinonlinear parameterization provides a good correlation betwcen true and parameterized cloud radiative fluxes. Although the quality of this parameterization may partly be due to the limited number of cloud realizations used in this study, the results suggest that domain averaged solar radiative fluxes may in principle be parameterized in terms of average cloud properties.