Abstract
Abstract. The frequent presence of cloud cover in polar regions limits the use of the Moderate Resolution Imaging Spectroradiometer (MODIS) and similar instruments for the investigation and monitoring of sea-ice polynyas compared to passive-microwave-based sensors. The very low thermal contrast between present clouds and the sea-ice surface in combination with the lack of available visible and near-infrared channels during polar nighttime results in deficiencies in the MODIS cloud mask and dependent MODIS data products. This leads to frequent misclassifications of (i) present clouds as sea ice or open water (false negative) and (ii) open-water and/or thin-ice areas as clouds (false positive), which results in an underestimation of actual polynya area and subsequently derived information. Here, we present a novel machine-learning-based approach using a deep neural network that is able to reliably discriminate between clouds, sea-ice, and open-water and/or thin-ice areas in a given swath solely from thermal-infrared MODIS channels and derived additional information. Compared to the reference MODIS sea-ice product for the year 2017, our data result in an overall increase of 20 % in annual swath-based coverage for the Brunt Ice Shelf polynya, attributed to an improved cloud-cover discrimination and the reduction of false-positive classifications. At the same time, the mean annual polynya area decreases by 44 % through the reduction of false-negative classifications of warm clouds as thin ice. Additionally, higher spatial coverage results in an overall better subdaily representation of thin-ice conditions that cannot be reconstructed with current state-of-the-art cloud-cover compensation methods.
Highlights
Information on cloud presence is of crucial importance when using thermal-infrared imagery
For a later comparison based on cloud coverage and polynya area, we extracted and use ice-surface temperature (IST) from the reference NSIDC MOD/MYD29 sea-ice product produced from Moderate Resolution Imaging Spectroradiometer (MODIS) Collection 6 data, which offers an overall accuracy of 1–3 K under ideal conditions (Hall et al, 2004; Riggs and Hall, 2015)
We describe and discuss the results from using our open-water–sea-ice–cloud discrimination (OSCD) product in comparison to the reference MOD/MYD29 seaice product on the basis of a thin-ice thickness (TIT) estimates (i) on a swath-to-swath basis, (ii) on the basis of daily composites of all available swaths per day, and (iii) as a comparison of overall achieved coverage over a year (Fig. 2f)
Summary
Information on cloud presence is of crucial importance when using thermal-infrared imagery. Several studies use MODIS thermal-infrared (TIR) data to monitor polynya area and associated sea-ice production in polynyas both in the Arctic as well as the Antarctic and compare well to or even outperform studies using passive-microwave satellite data in certain regions (e.g., Paul et al, 2015; Aulicino et al, 2018; Preußer et al, 2019). These studies generally utilize ice-surface temperature from the National Snow and Ice Data Center (NSIDC) sea-ice product The data set is derived using a combined approach of unsupervised deep learning, subsequent clustering, and manual screening from co-located 1 km resolution MOD/MYD02 product data (MODIS Characterization Support Team (MCST), 2017a, b) accessed through the Level-1 and Atmosphere Archive & Distribution System (LAADS) Distributed Active Archive Center (DAAC) and the Sentinel A/B (S1-A/B) synthetic aperture radar (SAR) calibrated backscatter data accessed through the Alaska Satellite Facility (ASF) DAAC as a cloud-independent reference
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call