It's almost completely dark save for the occasional streaks of lightning across a stormy sky. The waves would be at least three times Merrin's height if she were standing upright, which she is super not doing right now.
She is, instead, ducking under the water, which is cold but not that close to freezing, and it wouldn't bother her anyway because she's wearing a wetsuit.
(Which doubles as a surprisingly convincing aquatic-phase Sparashki cosplay, since apparently certain inside jokes never die.)
She's been vaguely on call for the last five hours; storms are high risk, prediction markets were saying it was likely something would go wrong, and policy markets bid for Merrin over the alternatives, largely because her particular Exception Handling training path has focused so heavily on operating alone or in a small group, for potentially long periods, without much backup. For a speculative investment, it's a lot cheaper to have a small team preemptively on call than a larger one.
(What Merrin is currently unaware of is that it is, actually, quite expensive to bid for Merrin's time. She has no idea how expensive because she totally took the ill-advised advice and has not looked at her own finances since she accepted the role.)
Even at the point when she was frantically summoned on an emergency helicopter an entire forty-five minutes ago for some sort of small boat related Problem, Merrin didn't really expect she would actually have to do all that much. Even then, unspecified problem doesn't mean 'something that needs Merrin's level of training to handle' - there are other ships in this general patch of ocean, cargo or scientific research vessels, big enough not to be themselves at risk and also big enough to have their own resources to contribute. In fact, one of them, started detouring their-way as soon as the Problem was officially reported - it's still almost 30 miles away, but could have adequately handled 'damaged small boat without power or steering plus exposure-related injuries.'
Usually when the emergency-forecasting prediction markets give a 5% chance of "something happening where it's worth a lot of money to have Merrin, specifically, on the response team", the thing doesn't in fact happen, as a basic premise of how probabilities work.