A neural network based forward operator for visible satellite images and its adjoint. (November 2021)

• Record Type: Journal Article
• Authors: Scheck, Leonhard
• Is Part Of: Journal of quantitative spectroscopy & radiative transfer. Volume 274(2021)
• Publication Date: 2021-11
• Subjects: Forward operator -- Satellite images -- Visible spectrum -- Data assimilation -- Machine learning -- Neural network -- Adjoint; Spectrum analysis -- Periodicals; Radiation -- Periodicals; Analyse spectrale -- Périodiques; Rayonnement -- Périodiques; Radiation; Spectrum analysis; Periodicals; 543.0858
• DOI: 10.1016/j.jqsrt.2021.107841
• Languages: English
• ISSNs: 0022-4073

Abstract

Highlights: Satellite images can be assimilated to improve numerical weather forecasts. Standard radiative transfer methods for the visible spectral range are too slow. A new neural network based method is very fast and sufficiently accurate. It requires orders of magnitude less training data than previous methods. The adjoint code of the network is fast and universally usable. Abstract: The benefits of using a machine learning approach in a forward operator for visible satellite images are explored. In the conventional version of the forward operator, cloud-affected reflectances are determined by linear interpolation in a compressed, seven-dimensional look-up table (LUT) computed with standard radiative transfer (RT) methods. It is demonstrated that replacing the LUT by a feed-forward neural network can reduce the computational effort by an order of magnitude without detrimental impact on the accuracy of the method. The sensitivity of the mean reflectance error to parameters controlling the network structure and the training process is investigated. Best results are obtained for networks with between four and eight hidden layers. Moreover, for the training of the network only 1 / 1000 of the data that has to be computed for the LUT using slow standard RT methods is required. The amount of memory required while generating synthetic images is reduced by a similar factor, compared to the LUT-based approach. The reduced requirements and increased speed strongly enhance the extensibility of the method. Adding more input parameters to account e.g. for traces gases, aerosols or more details in the cloud structure would be problematic for the conventional approach due to strongly increasing LUT sizes, but should be feasible in the neural network based version. A neural network inference code including tangent linear and adjoint versions was implemented to demonstrate further advantages of the new approach. In contrast to the LUT-based approach the derivatives computed with the adjoint of the neural network are continuous. Moreover, the adjoint code will not have to be changed when the network is trained with improved RT methods. The effort to keep the adjoint code in sync with the nonlinear code can thus be avoided.