Articles | Volume 10, issue 9
Atmos. Meas. Tech., 10, 3547–3573, 2017
https://doi.org/10.5194/amt-10-3547-2017
Atmos. Meas. Tech., 10, 3547–3573, 2017
https://doi.org/10.5194/amt-10-3547-2017

Research article 29 Sep 2017

Research article | 29 Sep 2017

Cirrus cloud retrieval with MSG/SEVIRI using artificial neural networks

Johan Strandgren et al.

Data sets

CALIPSO/CALIOP Level 2, Lidar Cloud Layer Data, version 3.01 CALIPSO Science Team https://doi.org/10.5067/CALIOP/CALIPSO/CAL_LID_L2_05kmCLay-Prov-V3-01_L2-003.01

CALIPSO/CALIOP Level 2, Lidar Cloud Layer Data, version 3.02 CALIPSO Science Team https://doi.org/10.5067/CALIOP/CALIPSO/CAL_LID_L2_05kmCLay-Prov-V3-02_L2-003.02

CALIPSO/CALIOP Level 2, Lidar Aerosol Layer Data, version 3.01 CALIPSO Science Team https://doi.org/10.5067/CALIOP/CALIPSO/CAL_LID_L2_05kmALay-Prov-V3-01_L2-003.01

CALIPSO/CALIOP Level 2, Lidar Aerosol Layer Data, version 3.02 CALIPSO Science Team https://doi.org/10.5067/CALIOP/CALIPSO/CAL_LID_L2_05kmALay-Prov-V3-02_L2-003.02

Download
Short summary
The new algorithm CiPS is presented and validated. CiPS detects cirrus clouds, identifies opaque pixels and retrieves the corresponding optical thickness, cloud top height and ice water path from the geostationary imager MSG/SEVIRI. CiPS utilises a set of four artificial neural networks trained with space-borne lidar data, thermal MSG/SEVIRI observations, model data and auxiliary data. To demonstrate the capabilities of CiPS, the life cycle of a thin cirrus cloud is analysed.