Getting Driving Directions from the BlueBoat

Happy New Year everyone. While I took time off from work for our winter vacation, I returned this week to finish putting together data from our December field excursions. Across 4 days in the field, we collected data for 10+ hours, all of which start as an organized mess of numbers. Our nitrate sensor and water quality sensor aboard the BlueBoat collect data every minute and each sensor records their collection time, which is really useful. When we deployed the BlueBoat for its missions, we had a team member record the start of the mission and end of the missions so that we could match the data. If we hadn't recorded these time stamps, we might struggle to determine which data are worth keeping and which data were from non-target locations, since both the nitrate sensor and water quality sensor collect data as soon as we turn them on, meaning they collect data before we put them in the water and after we retrieve them. 

Unfortunately, the BlueBoat adds a layer of mess to the data cleaning process. The BlueBoat at the time of writing records data every 0.1 seconds, but the BlueBoat does not have a real time clock. Instead, it logs the data as time since the boat was booted up. Aligning the BlueBoat GPS data with the water quality data, therefore, means finding the time stamp for the start of the mission and moving in 1 minute intervals to align these coordinates with the water quality data. If you're doing the math, that means filtering out 599 rows of data times the length of the excursion. Now, to find the start of the mission, I can upload the mission log to the UAV (unmanned aquatic/aerial vehicle) log viewer, which generates an image like the one you see here. I can trace along the graph and the BlueBoat icon will move along the path, and the system has the path color coded based on the driving mode. For our December missions, we piloted mostly using Auto mode (marked here by the yellow graph sections), so I can easily find the starting point and hover over the area of the graph to get the time stamp. In this process, though, I learned that if we are not thoughtful about how we work with the BlueBoat in the field, our data processing steps get longer, as the BlueBoat will record data whether its in the water or not. As long as it is on, the BlueBoat extends the log, which means more rows of data to sort through. Moving forward, we will hopefully remember to only turn the BlueBoat on when we are ready to deploy and that we will turn the BlueBoat off before we get it back onboard. 

We are planning our next set of missions for February and I'm hoping that everything we've learned from this first trip will lead to more success in the coming months. I'm especially interested in getting our BlueBoat super user-friendly, so that more members can participate in the various aspects of deployments. Stay tuned for more adventures!

---