Articles | Volume 10, issue 12
Atmos. Meas. Tech., 10, 4747–4759, 2017
https://doi.org/10.5194/amt-10-4747-2017
Atmos. Meas. Tech., 10, 4747–4759, 2017
https://doi.org/10.5194/amt-10-4747-2017

Research article 05 Dec 2017

Research article | 05 Dec 2017

Feasibility study of multi-pixel retrieval of optical thickness and droplet effective radius of inhomogeneous clouds using deep learning

Rintaro Okamura1, Hironobu Iwabuchi1, and K. Sebastian Schmidt2 Rintaro Okamura et al.
  • 1Center for Atmospheric and Oceanic Studies, Graduate School of Science, Tohoku University, 6-3 Aoba Aramaki-aza, Sendai, Miyagi, 980-8578, Japan
  • 2Department of Atmospheric and Oceanic Sciences, University of Colorado, Boulder, CO, USA

Abstract. Three-dimensional (3-D) radiative-transfer effects are a major source of retrieval errors in satellite-based optical remote sensing of clouds. The challenge is that 3-D effects manifest themselves across multiple satellite pixels, which traditional single-pixel approaches cannot capture. In this study, we present two multi-pixel retrieval approaches based on deep learning, a technique that is becoming increasingly successful for complex problems in engineering and other areas. Specifically, we use deep neural networks (DNNs) to obtain multi-pixel estimates of cloud optical thickness and column-mean cloud droplet effective radius from multispectral, multi-pixel radiances. The first DNN method corrects traditional bispectral retrievals based on the plane-parallel homogeneous cloud assumption using the reflectances at the same two wavelengths. The other DNN method uses so-called convolutional layers and retrieves cloud properties directly from the reflectances at four wavelengths. The DNN methods are trained and tested on cloud fields from large-eddy simulations used as input to a 3-D radiative-transfer model to simulate upward radiances. The second DNN-based retrieval, sidestepping the bispectral retrieval step through convolutional layers, is shown to be more accurate. It reduces 3-D radiative-transfer effects that would otherwise affect the radiance values and estimates cloud properties robustly even for optically thick clouds.

Download
Short summary
Three-dimensional (3-D) radiative transfer effects are a major source of retrieval errors in satellite-based optical remote sensing of clouds. Multi-pixel, multispectral approaches based on deep learning are proposed for retrieval of cloud optical thickness and droplet effective radius. A feasibility test shows that proposed retrieval methods are effective to obtain accurate cloud properties. Use of the convolutional neural network is effective to reduce 3-D radiative transfer effects.