Research Article
Adjusting Neural Network to a Particular Problem: Neural Network-Based Empirical Biological Model for Chlorophyll Concentration in the Upper Ocean
Table 1
Neural network inputs and outputs.
| Input # | Variable | Units | Input | Size |
| 1 | Year | | yr | 1 |
| 2 | Day of the year | | | 1 |
| 3 | Day of the year | | | 1 |
| 4 | Longitude | | | 1 |
| 5 | Longitude | | | 1 |
| 6 | Latitude | | | 1 |
| 7 | Sea surface temperature [SST] | °C | SST | 1 |
| 8 – 14 | Argo salinity [S(z)] | PSS | sal | 7 |
| 15 – 21 | Argo temperature [T(z)] | °C | temp | 7 |
| 22 | Ocean depth at current Lat, Lon | m | depth | 1 |
| 23 | Sea surface height [SSH] | m | SSH | 1 |
| Total | All Inputs | | | 23 |
| Output # | Variable | Units | Output | Size |
| 1 | Chlorophyll-a | mg/m3 | ln(chl-a) | 1 |
| 2 | | | ln() | 1 |
|
|