Using MODIS data to estimate sea ice thickness in the Bohai Sea (China) in the 2009–2010 winter

Abstract

To estimate sea ice thickness over a large spatial scale is a challenge. In this paper, we propose a direct approach to effectively estimate sea ice thickness over a large spatial area of the Bohai Sea using EOS MODIS data. It is based on the model of an exponential relation between albedo and thickness of sea ice. Eighteen images of EOS MODIS L1B data in the 2009–2010 winter were used to estimate the sea ice thickness and to monitor its spatiotemporal evolution in the Bohai Sea. The estimated thickness results are in accordance with results based on the Lebedev and Zubov empirical models as well as the forecasting data from the National Marine Environmental Forecasting Centre of China. Model correlation coefficients (R2= 0.864 and 0.858) and close similarity in thickness prediction attest to the reliability and applicability of the proposed method. The average ice thickness of the whole Bohai Sea ranged from 3 to 21 cm, with an estimated maximum about 40 cm in Liaodong Bay. Multiple‐temporal maps of sea‐ice thickness show that the sea ice formed initially along the coastline, and gradually expanded away from the shore. Sea ice first appeared in the Liaodong Bay, and hugged the coast southwards to Bohai and Laizhou Bay. During melting the inverse sequence occurred. Our results also show that sea ice coverage and thickness are significantly correlated with the value ofθ, the difference between cumulative FDD (Freezing Degree Days) and TDD (Thawing Degree Days).