DINCAE 2.0: multivariate convolutional neural network with error estimates to reconstruct sea surface temperature satellite and altimetry observations

Barth, Alexander; Alvera-Azcárate, Aida; Troupin, Charles; Beckers, Jean-Marie

doi:https://doi.org/10.5194/gmd-15-2183-2022

Articles | Volume 15, issue 5

https://doi.org/10.5194/gmd-15-2183-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-15-2183-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 15, issue 5

Development and technical paper

|

15 Mar 2022

Development and technical paper |

| 15 Mar 2022

DINCAE 2.0: multivariate convolutional neural network with error estimates to reconstruct sea surface temperature satellite and altimetry observations

Alexander Barth, Aida Alvera-Azcárate, Charles Troupin, and Jean-Marie Beckers

Data sets

MODIS Terra Level 3 SST Thermal IR Daily 4km Nighttime v2014.0, Ver. 2014.0 OBPG https://doi.org/10.5067/MODST-1D4N4

NOAA Optimum Interpolation 1/4 Degree Daily Sea Surface Temperature (OISST) Analysis, Version 2 R. W. Reynolds, V. F. Banzon, and NOAA CDR Program https://doi.org/10.7289/V5SQ8XB5

Model code and software

gher-ulg/DINCAE.jl: v2.0.0 (v2.0.0) A. Barth https://doi.org/10.5281/zenodo.6342276

Short summary

Earth-observing satellites provide routine measurement of several ocean parameters. However, these datasets have a significant amount of missing data due to the presence of clouds or other limitations of the employed sensors. This paper describes a method to infer the value of the missing satellite data based on a convolutional autoencoder (a specific type of neural network architecture). The technique also provides a reliable error estimate of the interpolated value.