This Russian Akula SSN is moving from right to left. The spikes towards the front of the fin/sail are probably wake detection sensors. These are known, in Russian, as System Obnarujenia Kilvaternovo Sleda (SOKS). SOKS was first tried in 1969 on Russia’s November class SSN K-14. Later types of SOKS may have been named Colossus, Toucan, and Bullfinch. These may have been mounted on all subsequent Soviet/Russian SSNs, including the curret Akulas and Yasens.
From the mid 1950s onward submarine detectors such as “Autolycus “ (ie. aircraft mounted sensors to “sniff” submarine diesel exhaust) and search radar were being developed against the snorkels of diesel subs. These became less effective with the shift of major powers to nuclear submarines from the 1960s. Nuclear subs could run submerged, without needing to snorkel. Occasional periscope use still left SSNs vulnerable to sensitive radar.
In Russia and the West, new methods were sought to detect submerged nuclear subs. Some of these methods detected not the submarine itself, but the disturbances it made in the sea. If a sub's motion mixed layers of cold surface and underlying warm water, this would raise the apparent surface temperature slightly.
This temperature change might be detected from infrared thermometers mounted on aircraft, UAVs, satellites. But satellites suffered from being stationary or following predictable paths. Another platform might be the US X-37 "uncrewed space plane" whose manoeuvrability could "surprise" Chinese or Russian nuclear submarines.
It may or may not be possible to detect a submarine by only measuring the temperature of the wake. Greater definition might be achieved by imaging the temperature of the surrounding sea. Such imaging might happen in real time, but may be assisted by preliminary studies of sea conditions using massive amounts of data (say around China's Hainan Island and Russia's Pacific SSBN Base at Rybachiy on the Kamchatka Peninsula).