Dust Aerosol Optical Properties Retrieval And Satellite Monitoring Of Duststorm Over Northwestern China

Dust aerosols, as a major component of aerosols in the atmosphere, can modulate the radiative energy balance of earth-atmosphere system and thus has large effect on the climate. The dust aerosols, which are blown up by strong wind, could also cause a severe havoc to environment and human activities. It is therefore important to fully understand dust aerosol radiative forcing effects on climate and monitor the evolution of dust storm. We developed a method to retrieve dust aerosol optical properties using the ground-based MFRSR measurements. We analyzed the data obtained from the 2008 China-U.S. joint field experiment and retrieved the optical properties of dust aerosols by applying this method. To validate satellite retrievals over Northwest China where the surface is relatively bright, we compared the aerosol optical depth (AOD) products from different satellites/retrieving methods with the AOD from the ground-based observations. We also examined the capability of microwave channels onboard satellite to detect dust storm and propose an integrated method to monitor dust using visible, infrared and microwave channels from the satellite.Several advanced instruments including MFRSR and CIMEL were deployed in the 2008 China-U.S. joint field experiment. We used the data from the Zhangye site to retrieve dust aerosol optical properties over Northwestren China. The AOD, which we derived from the MFRSR, is consistent with that from CIMEL. The AOD difference between MFRSR and CIMEL is less than 0.02 at 0.67μm wavelength. By assuming a spherical shape of dust particle, we iterate the parameters of size distribution until the RMS differences among the AOD values derived from the observations at five MFRSR channels and those from Mie calculations is minimum. It is found that the mean values of the fine and coarse mode radii are 0.137μm and 2.22μm, respectively. The volume concentration of large particles is 10 times of small particles. We then use the diffuse-to-direct ratio method (DDR) to retrieve the single-scattering albedo (SSA) and asymmetry factor (ASY). The values of SSA, which range from 0.76 at 0.415μm to 0.86 at 0.870μm, are much lower than those derived in Africa and also relatively smaller than those obtained over East Asian. Our results clearly show that the dust aerosol over Northwest China is much more absorptive. We further carry out a radiative closure experiment. The daylight-averaged differences between model and observations are-2.07 Wm-2 for the direct normal flux,-2.90 Wm-2 for the diffuse flux, and-8.54 Wm-2 for the total flux. The good agreement between simulations and measurements indicates that our retrieved dust optical properties are reasonable and reliable.Since the MFRSR measures the irradiances, the MFRSR retrievals may not be much sensitive to the shape of aerosol particles. We thus assumed that the shape of dust particles is spherical. However, the shapes of dust particles are irregular. In order to test our assumption, we calculate SSA, ASY, extinction efficiency (Qe) and scattering phase function for non-spherical particles using the T-matrix and IGOM methods and spherical dust particles using the Mie code. The differences of SSA, ASY and Qe between spherical and non-spherical particles are very small; however, the scattering phase function of spherical particle is quite different from that of non-spherical particle. Using all these dust optical properties along with the SBDART model to calculate the solar radiative flux reaching at surface, it is found that the solar radiative flux is not sensitive to different particle shapes. Therefore it is reasonable to retrieve dust aerosol optical properties by assuming a spherical shape. We also wrote a Monte Carlo radiative transfer code to examine the AOD error caused by the forward scattering of dust particles. We show that when the AOD values are less than 1, the AOD relative error caused by forward scattering is less than-3%. When the AOD value is greater than 1 with high solar zenith angle (＞60°), the relative error can reach-40%.Satellite observations are an effective way to provide a regional coverage of aerosol optical depth and monitor the occurrence and evolution of dust storm. We use surface retrieved AOD to validate the MISR and MODIS AOD products over Northwest China where the surface is bright. Our results suggest that MODIS retrieved AOD values are not reliable over semi-arid and arid scenes. This is most likely because the MODIS operational aerosol retrievals over land use the dark-target approach. In this approach the near-infrared (2.1 and 3.8μm) channels are employed to estimate the spectral surface reflectance in order to separate the surface and atmospheric components of the radiance received by the satellite. It also assumes an empirical relationship to deduce surface reflectances at wavelengths of 0.47 and 0.66μm using the remote sensed surface reflectance values at 2.1μm. The AOD based on the Deep Blue algorithm are much improved over the MODIS values because Deep Blue employs two blue channels (0.412 and 0.470μm) in MODIS, for which surface reflectances are relatively small, to infer aerosol properties. MISR provides radiance measurements of the same target at nine different viewing angles. Since MISR can remove the atmospheric path contribution from the surface-leaving radiance by taking advantage of differences in multi-angular signatures, MISR is much less sensitive to surface type and can successfully retrieve AOD over bright surfaces.The most common dust storms in East Asia are those caused by strong winds behind a cold front and generally coexist with cirrus. For example, the dust storm is especially often stirred up by Mongolian cyclone. The visible and IR techniques thus can not detect dust which is often under cirrus clouds. The microwave is not significantly attenuated by ice clouds and can penetrate ice cloud. We use AMSR-E and a microwave radiative transfer model to analyze dust effects on microwave. Both satellite measurements and model simulations show that dust particles can significantly reduce microwave radiation through the scattering and absorption effects. We propose a new method by defining a microwave polarized index (MPI) to detect dust storm underneath cirrus cloud. We then develop an ntegrated method to detect dust storm by combining the visible, infrared and microwave satellite instruments.