Over the past week, I have been looking at a way to check the accuracy of the tools we use to pilot our long duration gliders. So far, we have been using wind mapping sensors and wave height analysis from Marinemet, NCOM geostrophic currents, myocean currents at various depths through the water column, and the ROTFS model. This week I was able to run a preliminary analysis on the myocean and ncom data sets.
As both data sets are available in netcdf format, I was able to dissect them and easily pull out the data for the area around the glider. While the NCOM data is the geostrophic average, there is only one data point to directly analyze, but with myocean, we are able to take a number of points throughout the water column from surface to the depths Silbo dives to (in this case 1000m) to average. The myocean data is pretty interesting to compare, as in order to do so, I take a number of points and take the average, similar to how we arrive at the currents that Silbo records.
At the beginning of the week, I calculated the the % error of the values for the models using the numbers recorded by Silbo as the accepted value, and both were roughly within 15% from the magnitude of Silbo, and 20% when direction was compared.
Towards the end of the week I began making these plots that I will keep up when I have the chance, visualizing the magnitude and direction or the currents, withe the blue arrow being of the myocean data set, green being NCOM and red being Silbo.
Force Wind Sea & Honor
Tags: Nilsen Strandskov