A few years ago we were doing a project with NOAA to build a fixed wing ocean survey drone. (Our concept was to launch into the wind over the railing and then land in the water next to the ship.) We learned quickly that it is really hard to hold a fixed gps position with a large ship at sea. It can be done with a lot of effort, but the captain basically said nope. By the time we powered up and were ready to launch, the ship could have already drifted a mile (between current and wind) which was outside of our line-of-sight limitations. We were developing the system from scratch so I came up with the concept of a pattern route that would automatically relocate itself relative to the ship (errr ground station) position. The operator station had it's own gps and would relay it's position back up to the airplane every few seconds. We were able to demo the system successfully off the north shore of Oahu (from a rented dive boat.) Obviously what we were trying to do is totally different from your mission, but I just wanted to comment that there are many challenges to operating at sea ... some of those challenges can be a bit unexpected if it's your first time out there. (And I'm no ocean expert, I'm from land-locked MN.)
You might have some challenges calibrated your magnetometers on a ship with all that metal? There can be incredibly tricky winds in and around the decks of a ship, not to mention many obstacles like railings and lines and antennas. Everything is in constant perpetual motion. I don't ever get puking sea sick, but I do get a slight shade of green if I don't keep my eyes up on the horizon ... and when you go eyes down on your ground station for 10 minutes to draw a route, that can get a bit dicey when there is a lot of motion. Every location and day is different, but we were consistently getting 15-25 kt winds out on the unprotected ocean. I came away from the experience with a new respect for all of those that live and work on the ocean every day.
My best guess is that you'll need to be flying the mavic 100% manually and you won't ever have a chance to take your hands off the sticks or even get a break to hover with sticks centered -- even landing will end up needing to be in some amount of forward flight. Also, you probably want to bring at least 2 mavics in case the first flights are learning experiences. You'll definitely want to coordinate with the captain (I can't imagine they would be pleased with someone flying a drone from their ship without coordination with the bridge.) And you'll want to coordinate with the captain anyway to see if they'd be willing to idle the ship and hold a fixed heading during the flight. It might be a new experience for both of you, so good to be able to work together on it.
I know others have managed to make this sort of thing work so it has to be something that is doable.
For our fixed wing system, we had our fair share of issues. The final demo flight ended in disaster when a wing servo stopped functioning 42 seconds into the flight. We were able to figure out what happened from the low resolution flight log captured on the ground station, and then also were able to recover the shredded wreckage and got the full data log which was nice to have. Here was the first few glorious seconds out by the channel islands on a nearly dead calm day:
A few minutes after that video we were all standing around the shredded soggy wreckage on deck asking the obvious question: So how many can we sign you up for?