Investigating Bidirectional Reflectance in the Los Angeles Megacity Using CLARS Multiangle and Hyperspectral Measurements

The surface reflectance is a key ingredient in the remote sensing of surface and atmospheric properties from space. The determination of atmospheric composition, including greenhouse gas (GHG) and aerosol concentrations, from reflected sunlight requires accurate knowledge of the contribution from the underlying surface. Over megacity areas, such as the Los Angeles (LA) basin, which are major sources of GHGs and anthropogenic aerosols, the quantification of surface reflectance is challenging due to the associated complex land use types. In this study, we investigate the bidirectional reflectance in the Los Angeles megacity area using multiangle and hyperspectral radiance measurements from the California Laboratory for Atmospheric Remote Sensing (CLARS). The CLARS facility is located near the top of Mt. Wilson, at an altitude of 1670 m a.s.l., overlooking the LA megacity area with an FTS operating since 2011 to continuously monitor the GHGs and near-surface aerosols in the basin. The CLARS-FTS offers continuous high-resolution spectral measurements in the visible, near infrared and shortwave infrared spectral regions. The CLARS measurements mimic the off-nadir viewing of a low-Earth orbiting instrument, such as GOSAT and OCO-2, but with daily viewing capability. Eight surface targets with different land use types, including urban parks, industrial and residential areas, are selected in this study. The surface reflectance for specific solar incident and viewing angles is calculated by dividing, for non-absorbing spectral channels on clear days (such that gas and aerosol extinction can be ignored), the observed radiance reflected from surface targets by the observed irradiance. The non-linear Rahman-Pinty-Verstraete (RPV) model is used to model the Bidirectional Reflectance Distribution Function (BRDF) by fitting the multiangle and hyperspectral measurements. By evaluating the retrieved RPV parameters, we find that the RPV model provides a good representation of the BRDF in the LA megacity area. The fitted RPV parameters and their dependence on wavelength provides quantification of BRDF and potentially contributes towards reducing uncertainties in retrievals of GHGs and aerosols in megacity from space.