Fast Radiative Transfer Approximating Ice Hydrometeor Orientation and its Implication on IWP Retrievals
Artikel i vetenskaplig tidskrift, 2022

The accurate simulation of microwave observations of clouds and precipitation are com-putationally challenging. A common simplification is the assumption of totally random orientation (TRO); however, studies have revealed that TRO occurs relatively infrequently in reality. A more appropriate assumption is that of azimuthally random orientation (ARO), but so far it has been a com-putationally expensive task. Recently a fast approximate approach was introduced that incorporates hydrometeor orientation into the assimilation of data from microwave conically scanning instruments. The approach scales the extinction in vertical (V) and horizontal (H) polarised channels to approximate ARO. In this study, the application of the approach was extended to a more basic radiative transfer perspective using the Atmospheric Radiative Transfer Simulator and the high-frequency channels of the Global Precipitation Measurement Microwave Imager (GMI). The comparison of forward simulations and GMI observations showed that with a random selection of scaling factors from a uniform distribution between 1 and 1.4–1.5, it is possible to mimic the full distribution of observed polarisation differences at 166 GHz over land and water. The applicability of this model at 660 GHz was also successfully demonstrated by means of existing airborne data. As a complement, a statistical model for polarised snow emissivity between 160 and 190 GHz was also developed. Combining the two models made it possible to reproduce the polarisation signals that were observed over all surface types, including snow and sea ice. Further, we also investigated the impact of orientation on the ice water path (IWP) retrievals. It has been shown that ignoring hydrometeor orientation has a significant negative impact (∼20% in the tropics) on retrieval accuracy. The retrieval with GMI observations produced highly realistic IWP distributions. A significant highlight was the retrieval over snow covered regions, which have been neglected in previous retrieval studies. These results provide increased confidence in the performance of passive microwave observation simulations and mark an essential step towards developing the retrievals of ice hydrometeor properties based on data from GMI, the Ice Cloud Imager (ICI) and other conically scanning instruments.


radiative transfer

remote sensing

ice water path

polarisation differences


oriented hydrometeors


Inderpreet Kaur

Chalmers, Rymd-, geo- och miljövetenskap, Geovetenskap och fjärranalys

Patrick Eriksson

Chalmers, Rymd-, geo- och miljövetenskap, Geovetenskap och fjärranalys

Vasileios Barlakas

Chalmers, Rymd-, geo- och miljövetenskap, Geovetenskap och fjärranalys

Simon Pfreundschuh

Chalmers, Rymd-, geo- och miljövetenskap, Geovetenskap och fjärranalys

Stuart Fox

Met Office

Remote Sensing

2072-4292 (ISSN)

Vol. 14 7 1594

Mot nya satellitinstrument, för bättre väder-och klimatprognoser

Rymdstyrelsen (166/18), 2019-01-01 -- 2021-12-31.

Robust Estimation of Global precipitation using Neural networks (REGN)

Rymdstyrelsen (154/19), 2020-01-01 -- 2023-12-31.


Meteorologi och atmosfärforskning





Mer information

Senast uppdaterat