Summer Internship

Undergraduate research in University of California, Los Angeles

I worked with Professor Jasper Kok in the University of California, Los Angeles from 2018.7 to 2018.9

Publication

Zhang, Y., Huang, Y. and Kok, J. F. “Validation of the single-scattering properties of tri-  axial ellipsoidal dust against laboratory measurements”, manuscript in preparation



Mineral dust, one of the most widespread aerosol, primarily originating in desert and semi-arid regions. The abundance of dust in atmosphere can be as high as tens of millions of tons. In addition, dust aerosol interacts with clouds by serving as ice nucleus and indirectly impact the precipitation process, directly absorbs and scatters the radiation and contributes to the greenhouse effect and thereby altering the climate both locally and globally.

Remote sensing techniques, either ground-based ones (for example, Aerosol Robotic Network, AERONET; Dubovik et al., 2006) or space-based ones (for instance, Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations, CLIPOSO; Yang et al., 2012), use pre-calculated single-scattering properties of dust to retrieve dust information. As such, the uncertainty from these single-scattering properties will propagate to the remote-sensing retrievals, and further to simulation results by climate models. There are one of the main sources of error in dust aerosol property retrieval: the oversimplification of the morphological details of dust-like particles which assumes that all of the dust particles are spherical while most of dust are irregular particles in reality.

Remote sensing techniques, either ground-based ones (for example, Aerosol Robotic Network, AERONET; Dubovik et al., 2006) or space-based ones (for instance, Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations, CLIPOSO; Yang et al., 2012), use pre-calculated single-scattering properties of dust to retrieve dust information. As such, the uncertainty from these single-scattering properties will propagate to the remote-sensing retrievals, and further to simulation results by climate models. There are one of the main sources of error in dust aerosol property retrieval: the oversimplification of the morphological details of dust-like particles which assumes that all of the dust particles are spherical while most of dust are irregular particles in reality.

It is demonstrated that the tri-axial ellipsoid model is better than the the spheroid model for simulating the optical properties of nonspherical feldspar particles. In my study, I assume the dust aerosols are tri-axial ellipsoid particles and compute its single-scattering properties by interpolation. And I obtain the simulation result with Meng database then validate it by comparing my result with 6 dust samples from AGLSD.

Meng database is a combination of four generally accepted computational methods, including the Lorenz-Mie theory, the T-matrix method, the discrete dipole approximation, and an improved geometric optics method, covering various aspect ratios and size parameter range from Rayleigh to geometric optics, and used to compute the phase matrix, extinction efficiency, single scattering albedo and so on. (Meng et al. 2010)

The Amsterdam-Granada light scattering database (AGLSD; Muñoz et al., 2012) reports laboratory-observed single-scattering properties and the particle size distributions of six dust samples (named as feldspar, red clay, green clay, olivine L, Sahara Libya, and Sahara, respectively). Because of its comprehensiveness, AGLSD has been widely regarded as the standard to evaluate numerical models and algorithms assuming different dust shapes, such as spheres, spheroids, ellipsoids, Koch-fractal polyhedral, polyhedral prisms, hexahedra, Gaussian spheres and other shapes (Jin et al., 2016).