After a bit of search, I stumbled across two projects that relate to what I am trying to accomplish. The first being a fixed data logger that measures the conductivity which can be an indicator of water quality.
Pure water is a good insulator as the conductivity comes from the minerals that are dissolved within the water. So the idea here as that if the water quality goes down, i.e someone discharges waste into the water, the conductivity goes up. Temperature has also been integrated, which can be an indicator of abnormal input, such as flood water which is likely to be warmer or colder then what is expected to be normal. These two measurements are a general indicator as it does not tell you what exactly is in the water. Its also good because its low cost and easy to use as its built around an arduino. Downside is its built to hang off a bridge or a fixed object, so no sea for this idea, also the project seams to have stalled.
The next is open source weather buoy.
The idea is to create a 3d printable buoy that can be used to measure wave data using an accelerator. This is interesting as it is low cost and easy to build, but to calculate the difference the height of a wave may be difficult. Micro controllers have to run in a loop, which may cause an event to me missed and if an equation it may be too heavy for the controller to handle in a timely manor more information may be missed. Faster processors like that found in a raspberry pi require significantly higher amounts of power meaning that either you have to have bigger batteries or more power generation on board. The other problem is that your equation has to use an integral, which as time goes on, makes the data less accurate. Filters are then needed and so preference needs to be increased.
Alternatively you store the raw data for processing later, which may eat up your storage space. Other sensors such as using GPS data become unreliable as they can have as much as +/- 10m inaccuracy and more traditional methods would require a larger buoy as they measure movement of a rope, spring compression, movement of a large object, distance to sea floor using a laser or sonar which requires more power, dependent on water particulates and no change to sea bed, E.g Sea weed growing.
It a tough one to nail down what is important and whats not and how to go about it, more features, require more power and cost goes up, so it becomes difficult to strike a balance