Begin Main Content

Presenter: Lars Nerger

Seminar Title: Estimation of Model Bias by the Assimilation of Satellite Ocean Chlorophyll Data into a Global Model

Chlorophyll concentration estimates by ocean-biogeochemical models typically show significant errors. Data assimilation algorithms based on the Kalman Filter can be applied to improve the model state. However, these algorithms usually do not account for possible biases in the model prediction. Taking model bias explicitly into account can improve the assimilation estimates. Here, the effect of bias estimation is studied with the assimilation of chlorophyll data from the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) into the NASA Ocean Biogeochemical Model (NOBM). The ensemble-based SEIK filter has been combined with an online bias correction scheme. A static error covariance matrix is used for simplicity.

The performance of the filter algorithm is assessed by comparison with independent in situ data over the 7-year period 1998-2004. Compared to the assimilation without bias estimation, the bias correction results in significant improvements of the surface chlorophyll estimates. With bias estimation, the daily chlorophyll estimates from the assimilation show about 3.3% lower error than the SeaWiFS data. In contrast, the error in the global surface chlorophyll estimate without bias estimation is 10.9%.

return to Seminars