Summer snow plays an essential role in Arctic hydrology and in maintaining mass and energy balance of sea ice. However, there are great challenges in retrieving long‐term summer snow depths over Arctic sea ice. Here, we proposed a combined novel five‐variable long short‐term memory (hereafter CN5VLSTM) model based on brightness temperature data to yield warm‐season snow depth estimates. Then, year‐round snow depth estimates we...