I put energetics modelling work on hold for the latter part the fall semester (2018) to focus on getting oriented with Regional Ocean Modelling System (ROMS). ROMS is a grid-based primitive ocean equations hydrodynamic model that University of Rhode Island (URI) Associate Research Scientist Dave Ullman is working with for all the coastal waters of Rhode Island. For this aquaculture project, the goal is to predict the environmental variables across coastal RI needed to understand kelp and oyster growth. For example, for the kelp energetics model, we need a time series of dissolved inorganic nitrogen, total inorganic carbon, and irradiance. The reason that Dave started helping me learn ROMS is that in his bigger state-wide model the Saugatucket River is too small to be included in the set of rivers impacting the model. This absence matters for this aquaculture project because the Saugatucket River feeds into Point Judith Pond where two of our sites are located (Cedar Island Oysters).
This semester I am taking a great class at URI called Programming for Scientists with Professor Brice Loose, and a large part of this class is a final project where we utilize our coding skills in any scientific way that interests us. Brice decided to let me try to get as far as I could into learning how to run a small ROMS model for the Point Judith Pond area. The goal is to link this smaller model to the larger model that Dave is working on. A BIG thank you to Dave for all the time he has spent and continues to spend meeting with me to help me get oriented in ROMS world. Through meeting with Dave, I gained great appreciation for how much work goes into getting these models to run accurately. I would learn about a few file types and think I understand the whole picture and then be shocked the next week to discover that there were three more files to set up.
The main task that I have completed so far in ROMS world is figuring out how to generate a grid that ROMS can understand and run its equations along. There are multiple options in terms of programs that people have written to aid in grid generation, and I spent a bit of time working to download and consider different options. What ended up working for me is a code file in MATLAB called EASYGRID. Checkout the grid I created through the modification of this code (shown above)! We have the cells set at a pretty fine scale, so they almost look like a black fuzz across the image. More ROMS struggles to come!
Written by: Celeste Venolia