Ensemble forecast

The previous 'deterministic forecast' example does not take into account uncertainties in the initial condition (a.k.a. analysis) of the forecast model. A key advantage of SeaMon-API is that it provides easy access to advanced ensemble forecast products. Ensemble forecasts take into account multiple possible initial conditions, by executing multiple forecast runs with perturbations representative of the uncertainty of the initial condition. The following example fetches an ensemble forecast from the NCEP GEP model and compares with the deterministic NCEP GFS model:

ensemble forecast

https://api.seamon.io/forecast?waypts=(Thorlakshofn;2018-12-22T12:00),(Azores_Islands;2018-12-27T18:00)&params=gep|wind&view=graph

deterministic forecast side by side

The usually small uncertainties in the initial conditions of the forecast models naturally lead to growing uncertainty in prediction as you look further and further into the future. This is a completely natural and inescapable consequence of the chaotic nature of the Earth weather system. For this reason ensemble Monte Carlo approaches to the weather prediction are an extremely important for evaluating likely weather scenarios further than 3-4 days into the future.