Determination of the Predominant Orientation of Cracks in Ice Cover of the Laptev Sea from the Ice Drift Data

Abstract

Listen

Translate article

Received April 10, 2023; revised May 25, 2023; accepted June 27, 2023The paper shows that in order to solve the problem of predicting the predominant orientation of cracks in ice covering the Laptev Sea, when the satellite information is absent, the orientation of cracks in ice cover can be calculated using data on the ice drift in the Sea. We have selected 200 satellite images made in both visible and infrared spectral bands allowing to obtain actual data on ice cracks in the Laptev Sea for seasons 2016–2021, and then calculated the predominant orientation of them using data on the OSI SAF ice drift for the date of each image as well as for several previous days. Results obtained were compared with actual modal orientations of the cracks and with characteristics of the atmospheric pressure fields in these days. It was found that selecting of initial ice drift data should be done with regard for the intensity of air flows, the invariability or constancy of the direction of air flows over the sea aquatory during several days, and the presence of a cyclone centered over the sea area. An algorithm has been developed for automatic selection of a date when the ice drift can be most likely accompanied by formation of cracks on the day under consideration, and calculation of predominant orientation of cracks should be made exactly on that day. The selected data on ice drift serve as a basis for calculating. It was revealed also that with the difference in the types of intensity of air flows in the northern and southern parts of the Laptev Sea, the developed algorithm for determining the date of ice drift should be applied to each part of the sea separately. The number of squares of the Laptev Sea grid in which the calculated orientation of cracks corresponded to its actual meaning with an established permissible error of 30° has been determined. The ratio of this number to the total quantity of grid squares (skill score) was equal to 69% on the average. The skill score exceeded 70% in 46 percent of the cases. The skill score falls below 60% in only 16 percent of the cases, with minimum value of 50%.