Satellite cloudy radiances are rarely used in the data assimilation system because of the uncertainty in cloud properties predicted by NWP models. This study will investigate this uncertainty by simulating satellite radiances using a cloud model and comparing to the measurements from satellite visible sensors. Atmospheric and surface parameters generated from NWP models are used to simulate the response of various sensors to cloud and aerosol parameters (e.g. size and concentration). The simulations are compared with coincident measurements from current operational sensors such as AVHRR and MODIS.

From future sensors, the measurements of the Stokes vector at shorter wavelengths will be made in an effort to improve the remote sensing capabilities of aerosols, cloud particle size and shape. The U.S. National Polar-Orbiting Environmental Satellite System (NPOESS) developed a new sensor called aerosol polarimetry sensor (APS) that will provide the four Stokes component measurements at blue to near infrared bands and at multiple viewing directions. These Stokes vectors will provide us with unique signatures of aerosol and cloud particle size and shape. For example, the polarized reflectance near 140° scattering angle is significantly different for stratocumulus clouds than for cirrus clouds. Direct use of radiance measurements from APS in weather and climate models may be possible during the NPOESS era. This study investigates the use of a vector radiative transfer model as discussed in PART I (forward and adjoint model) under the major title for radiance and sensitivity calculations.