10 Nov 2020

10 Nov 2020

Review status: a revised version of this preprint is currently under review for the journal AMT.

Evaluation of VIIRS Neural Network Cloud Detection against Current Operational Cloud Masks

Charles H. White1, Andrew K. Heidinger2, and Steven A. Ackerman1 Charles H. White et al.
  • 1University of Wisconsin - Madison, Department of Atmospheric and Oceanic Sciences, Madison, WI, USA
  • 2NOAA/NESDIS/Center for Satellite Applications and Research, Madison, WI, USA

Abstract. Cloud properties are critical to our understanding of weather and climate variability, but their estimation from satellite imagers is a nontrivial task. In this work, we aim to improve cloud detection which is the most fundamental cloud property. We use a neural network applied to Visible Infrared Imaging Radiometer Suite (VIIRS) measurements to determine whether an imager pixel is cloudy or cloud-free. The neural network is trained and evaluated using four years (2016–2019) of coincident measurements between VIIRS and the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP). We successfully address the lack of sun glint in the collocation dataset with a simple semi-supervised learning approach. The results of the neural network are then compared with two operational cloud masks: the MODIS-VIIRS Continuity Cloud Mask (MVCM) and the NOAA Enterprise Cloud Mask (ECM).

We find that the neural network outperforms both operational cloud masks in most conditions examined with a few exceptions. The largest improvements we observe occur during the night over snow or ice covered surfaces in the high latitudes. In our analysis, we show that this improvement is not solely due to differences in optical depth-based definitions of a cloud between each mask. We also analyze the differences in true positive rate between day/night and land/water scenes as a function of optical depth. Such differences are a contributor to spatial artifacts in cloud masking and we find that the neural network is the most consistent in cloud detection with respect to optical depth across these conditions. A regional analysis over Greenland illustrates the impact of such differences and shows that they can result in mean cloud fractions with very different spatial and temporal characteristics.

Charles H. White et al.

Status: final response (author comments only)
Status: final response (author comments only)
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

Charles H. White et al.

Charles H. White et al.


Total article views: 292 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
197 92 3 292 9 8
  • HTML: 197
  • PDF: 92
  • XML: 3
  • Total: 292
  • BibTeX: 9
  • EndNote: 8
Views and downloads (calculated since 10 Nov 2020)
Cumulative views and downloads (calculated since 10 Nov 2020)

Viewed (geographical distribution)

Total article views: 259 (including HTML, PDF, and XML) Thereof 259 with geography defined and 0 with unknown origin.
Country # Views %
  • 1


Latest update: 03 Mar 2021
Short summary
Automated detection of clouds in satellite imagery is an important practice that is useful for predicting and understanding both weather and climate. Cloud detection is often difficult during at night and over cold surfaces. In this paper, we discuss how a complex statistical model (a neural network) can more accurately detect clouds compared to currently used approaches. Overall, our results suggest that our approach could result in more reliable assessments of global cloud cover.