Estimating oceanic chlorophyll concentrations with neural networks

A Neural network (NN) algorithm was constructed to estimate oceanic chlorophyll concentrations from Sea-viewing Wide Field-of-view Sensor (SeaWiFS) data.The algorithm was trained and tested using sample data compiled at the SeaWiFS Bio-optical Algorithm Mini-workshop. A NN using the five visible SeaWiFS bands as inputs with ten nodes in a single hidden layer estimates chlorophyll concentrations efficiently and accurately, modelling the nonlinear transfer function between surface chlorophyll concentrations and remotely sensed reflectance data more accurately than traditional regression methods.