Uncertainty quantification in the retrieval of cloud properties for Sentinel-4 and TROPOMI / Sentinel-5 Precursor (S5P) using deep neural networks
Abstract
In atmospheric remote sensing, the quantities of interest (e.g. cloud properties) are usually not directly observable but can only be inferred indirectly via the measured spectra. To solve these inverse problems, retrieval algorithms are applied that usually depend on complex physical models, so called radiative transfer models (RTMs). These are very accurate, however also computationally very expensive and therefore not feasible in combination with the strict time requirements of operational processing. Therefore, machine learning methods, in particular neural networks (NNs), can be used to accelerate the classical remote sensing retrieval algorithms. However, their application may be difficult as they can be used in different ways and there are many aspects to consider as well as parameters to be optimized in order to achieve satisfying results.
For the inverse problem in atmospheric remote sensing, there are two main approaches to apply NNs: - NNs used as forward model, where a NN approximates the radiative transfer model and can thus replace it in the inversion algorithm - NNs for solving the inverse problem, where a NN is trained to infer the atmospheric parameters from the measurement directly The first approach is more straightforward to apply. However, the inversion algorithm still faces many challenges, as the spectral fitting problem is generally ill-posed. Therefore, local minima are possible and the results often depend on the selection of the a-priori values for the retrieval parameters. For the second case, some of these issues can be avoided: no a-priori values are necessary, and as the training of the NN is performed globally, i.e. for many training samples at once, this approach is potentially less affected by local minima. However, due to the black-box nature of a NN, no indication about the quality of the results is available. In order to address this issue, novel methods like Bayesian neural networks (BNNs) or invertible neural networks (INNs) have been presented in recent years. This allows the characterization of the retrieved values by an estimate of uncertainty describing a range of values that are probable to produce the observed measurement. At the example of the S4 and S5P cloud product we apply and evaluate these new methods in order to demonstrate their potential for future operational algorithms.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFMIN33A..04R