NCEP's Global Ocean Data Assimilation System: a reanalysis that blends moorings, floats and satellites into a continuous three-dimensional field of the ocean, monthly, on a 1° by ⅓° grid with 40 depth levels.The value of such a synthesis lies in its completeness. Individual observations are snapshots in space and time, but GODAS fills the gaps using the laws of ocean physics, creating a continuous field that can be analyzed as a whole. This allows scientists to study large-scale phenomena like El Niño, track the transport of heat poleward, or monitor changes in ocean heat content. Because it is a reanalysis—a consistent reconstruction over many years—it also serves as a baseline for understanding long-term trends and for initializing seasonal climate forecasts, which depend on knowing the current state of the ocean to predict future atmospheric conditions.
GODAS was developed at the National Centers for Environmental Prediction (NCEP) in the early 2000s, building on earlier ocean reanalysis efforts. It became operational in 2003, providing a standard product for climate monitoring and seasonal prediction.
GODAS operates by combining a numerical ocean model with real observations through a process called data assimilation. The ocean model simulates the evolution of currents, temperature, and salinity based on physical equations, but its predictions drift from reality over time. Data assimilation corrects this drift by periodically pulling the model's state toward the actual measurements, weighting each observation according to its estimated error. The result is a model trajectory that is continuously nudged by reality, producing a physically consistent and observationally constrained estimate of the ocean.
GODAS is used to initialize seasonal climate forecasts, such as those predicting El Niño and its global impacts, and to monitor ocean heat content and sea-level changes for climate research.
The system's grid has roughly 1.4 million grid points per depth level, yet it relies on only a few thousand real-time observations to correct its state each day.
