Based on the daily NCEP-DOE Reanalysis II data from 1979 to 2018, the Fuzzy C-Means Algorithm (FCMA) was used to classify the climate during the Arctic polar day into four climate types, which were the cold-dry type, the semi-cold-dry type, the semi-warm-wet type, and the warm-wet type. Results show that under the climate background of rapid reduction of Arctic sea ice in summer, the regions controlled by these four climate types also change significantly. Among them, the area of the cold-dry type decreases rapidly with the rate up to -7.241×10⁴ km²·a^-1, the semi-cold-dry type increases at a rate of 6.489×10⁴ km²·a^-1,and the climate change on the ocean is more significant than that on the land. Around 1999, the dominant climate in the Arctic changed from cold-dry type to semi-cold-dry type, especially in the Chukchi Sea and East Siberian Sea. The genetic analysis shows that there is a close relationship between sea ice area, net short wave radiation, atmospheric warming and humidification, downward long-wave radiation as well as climate type transformation, moreover, the abnormal decreasing of sea ice area in polar day plays an important role in the abrupt change of climate transformation.