Nitrogen Oxide biogenic emissions from soils are driven by soil and environmental parameters. The relationship between these parameters and NO fluxes is highly non linear. A new algorithm, based on a neural network calculation, is used to reproduce the NO biogenic emissions in West Africa during the AMMA campaign, in August 2006. It has been coupled in the surface scheme of a coupled chemistry dynamics model to estimate the impact of the NO emissions on NO_x and O₃ formation in the lower troposphere. Four different simulations on the same domain and at the same period are compared: CTRL run (without soil NO emissions), YL95 run (with NO emissions inventory, at low time and space resolution), SOILNOx run (with NO emissions from neural network) and ALLNOx run (with NO from neural network). The influence of NO_x from lightning is assessed, and is limited to the upper troposphere. Compared to parameterisations generally used at the global and regional scales, the neural network parameterisation can give higher NO_x (up to +380 ppt) and ozone (up to +7ppb), closer to the ones measured in aircrafts during the AMMA field campaign. The NO emission from soils calculated with neural network responds to changes in soil moisture giving enhanced emissions over the wetted soil, as observed by aircraft measurements after the passing of a convective system, well reproduced by the model. Consecutive enhancement of NO_x and ozone is limited to the lowest layers of the atmosphere in modelling, whereas measurements show higher levels above 500 m. This equation allows an immediate response of fluxes to environmental parameters, on the contrary to fixed emission inventories. The annual cycle of emissions from this algorithm will be simulated in a future work