Journal of Quantitative Spectroscopy & Radiative Transfer 79–80 (2003) 757–764 www.elsevier.com/locate/jqsrt Discriminating between spherical and non-spherical scatterers with lidar using circular polarization: a theoretical study Yong-X. Hu a ; * , Ping Yang b , Bing Lin a , Gary Gibson a , Chris Hostetler a a MS 420, NASA LaRC, Hampton, VA 23681-2199, USA b Department of Atmospheric Science, Texas A& M, College Station, TX 77843, USA Received 1 June 2002; accepted 16 August 2002 Abstract For ground based observations, depolarization of lidar backscatter indicates that the scattering particles are non-spherical. This property provides a useful means to discriminate between ice particles (non-spherical) and water droplets (spherical) in clouds. However, for space based lidar measurements, backscatter from spherical water cloud particles is also depolarized due to multiple scattering. For the spaceborne lidar application, the discrimination between water and ice is not straightforward. An alternative method for water/ice discrimination that is less sensitive to multiple scattering is proposed in this study. The new approach is based on the dierences in P 44 (an element of the scattering phase matrix) at 180 ◦ scattering angle between spherical and non-spherical particles. By transmitting a circularly polarized beam from the lidar and resolving the rotational sense of the polarization in the receiver, discrim- ination between spherical and non-spherical scatterers can be accomplished even when multiple scattering occurs. When the incident beam is left-hand-circularly polarized, the circular component of backscatter by a non-spherical particle is weak and possibly left-handed, whereas backscatter by a spherical particle is sig- nicantly right-hand-circularly polarized. Monte Carlo simulations with full Stokes vector parameterizations indicate that multiple scattering does not aect the rotational sense of the backscatter polarization, making robust discrimination between spheres and non-spheres possible with this new circular polarization approach. Published by Elsevier Science Ltd. 1. Introduction Retrievals of optical thickness and eective particle size of a cloud from radiometric measurements depend critically on assumptions on the shape of the cloud particles. If cloud particle shape is unknown, uncertainties arise in the interpretation of the spectral and angular radiometric * Corresponding author. Tel.: +1-757-864-9824; fax: +1-757-864-7775. E-mail address: y.hu@larc.nasa.gov (Y.-X. Hu). 0022-4073/03/$ - see front matter. Published by Elsevier Science Ltd. doi:10.1016/S0022-4073(02)00320-5