On the timescales over which ocean circulation processes occur (i.e. up to ~1000 years), there are no significant sources of water within the deep ocean. However, what you're describing (plumes of low-density water) does occur at one particular setting, namely deep-sea hydrothermal vents. Extremely hot water (which can be well above 100C because the boiling point of water increases with pressure) enters the ocean at these vents and, because the water is so hot, it has a lower density and therefore rises up in a plume (e.g. see this figure, from acoustic imagery of a hydrothermal vent). The plume continues to rise until the plume water has mixed and cooled sufficiently to reach a neutral density. Note that the water exiting from hydrothermal vents isn't "new" water, it's primarily water that has either been circulating through fissures in the seafloor and heated up due to the high geothermal gradient and through proximity with melt pockets.