Characterisation of ocean forecast errors from an ocean forecasting system
Description of Work:
A state-of-the-art prediction system makes several assumptions about the errors of the observing system, the ocean models, the atmospheric forcing and data assimilation methodology. Correctly modelling and estimating these errors and validating or improving these assumptions is critical to further improving performance. This project will focus on the available database of forecast innovations and increments from the BLUElink ocean prediction system and determine the systematic bias as well as the statistical distribution. Specific methods will then be developed to deconstruct and attribute error to different components of the system as well as hypothesis testing.