Remote sensing of aerosols, clouds and trace gases using synergy of AATSR, MERIS, and SCIAMACHY onboard ENVISAT
Final Report Abstract
The main objective of this project was to develop new radiative transfer algorithms that enable the simulation of measurements of the remote sensing instruments AATSR, MERIS and SCIAMACHY on-board ENVISAT. The Monte Carlo model MYSTIC has been used as a basis. Before the project started, two important features were missing: (A) it was not possible to simulate measured radiances in presence of clouds without approximations regarding the scattering process and (B) spectra could not be simulated with sufficient accuracy for retrievals based on differential optical absorption spectroscopy (DOAS). During the project new algorithms were developed to overcome these problems. To be able to simulate scattering by cloud droplets and ice crystals without approximation, highly sophisticated variance reduction methods including so-called detector directional importance sampling, n-tuple local estimate, prediction-based splitting and Russian roulette have been developed and included in the MYSTIC code. The simulation of spectra of differential optical thickness was not possible because conventional Monte Carlo radiative transfer models are monochromatic, i.e. each wavelength is calculated separately resulting in independent statistical errors for each wavelength. The computational time becomes extremely large when the spectra need to be sufficiently accurate to resolve differential absorption features. This problem has been solved by calculating all wavelengths based on the same photon paths and consider the spectral variations by appropriate spectral weights. This new method called Absorption Lines Importance Sampling (ALIS) has been included in MYSTIC. It has been shown that it is very efficient for one- or three-dimensional radiative transfer in realistic model domains including clouds. It might even be fast enough to be used as forward model for DOAS based retrieval algorithms. The extension of the radiative tranfer code MYSTIC has been finalized so that it can be used for the generation of synthetic datasets for the model-based validation of retrieval algorithms to derive aerosol, cloud and trace gas properties from the remote sensing AATSR, MERIS and SCIAMACHY.
Publications
- ALIS: An efficient method to compute high spectral resolution polarized solar radiances using the Monte Carlo approach. J. Quant. Spectrosc. Radiat. Transfer, 112, 1622-1631, 2011
C. Emde, R. Buras, R., and B. Mayer
- Efficient unbiased variance reduction techniques for Monte Carlo simulations of radiative transfer in cloudy atmospheres: the solution. J. Quant. Spectrosc. Radiat. Transfer, 112, 434-447, 2011
R. Buras and B. Mayer
