Geologically speaking, today was very similar to many recent days, and included dredging, describing and cataloguing rocks from the ocean floor. However, the three of us, Daniel, Blake and I (Ryan) have been working to try and smooth out the much-heralded missing pinger dilemma. Several people on the boat have been helping us out as we try to model the behavior of the dredge as it moves through the water and is dragged behind the ship. Without the pinger, our technological advantage has been pushed back thousands of years, as we now have a bucket being dragged behind a ship, with no locating device on it. However, by modeling the physical characteristics of the dredge on the computer, in real time, with the ship’s position, and displaying the detail of the ocean’s underwater terrain, hopefully the scientists will be able to have a far better estimate of where the dredge is.

This program we are creating will have applications, even for those who are graced by the presence of modern technology (a pinger). It can tell possible future locations of the dredge, thanks to some other features already in effect. By using the waypoints from the dredge plan, on the computer, we are able to display what the underwater topography (bathymetry), should look like as we approach it. The area in front of the ship is what the bathymetry along the predicted path looks like, and behind the boat, is what we have experienced. When they mesh nicely (when not if), it increases the confidence we already have that what is upcoming on the ideal path is actually there, this allows the scientist dredging to see if they are about to become stuck on a little undersea ledge, or if they should let the dredge out further to touch the bottom.

Modeling the actual behavior of the dredge and dredge cable is far more complicated however, than displaying a 2-d version of the bathymetry in the ship’s path. Oddly enough, we are using many of the same methods that geologists use when trying to model tectonic forces, we first looked at all of the available data (dredge logs for the last week), and then tried to create a reasonable model based upon what we knew. We eventually found out, unfortunately, that our first attempt was not completely accurate, so we divined a prediction, then waited for the next dredge, and used the new data to refine our model as to make it more accurate. Essentially, we are creating the most plausible representation of the data which we currently have, when the prediction based on that model does not accurately predict future dredges, we are updating our model to reflect the new data.

With enough information, our program will eventually be able to predict the location of the dredge with a very high degree of accuracy, so perhaps it is not odd at all that we are on a ship here in Samoa collecting rocks, which will help geologists accumulate more data on the ‘hot spot’ model of plate tectonics. As we have learned through our own limited experience in trying to find out where the dredge is, the best way to make a model is to collect as much data as is possible, and then to revise and update that as necessary, exactly the thought process and reasoning behind this very oceanographic expedition to Samoa.

Ryan Delaney onboard the R/V Kilo Moana.