Importance of N2 fixation vs. nitrate eddy diffusion along a latitudinal transect in the Atlantic Ocean

We present ocean, basin-scale simultaneous measurements of N-2-fixation, nitrate diffusion, and primary production along a south-north transect in the Atlantic Ocean crossing three biogeographic provinces: the south subtropical Atlantic (SSA; similar to 31 degrees S-12 degrees S), the equatorial Atlantic (EA; similar to 12 degrees S-16 degrees N), and the north subtropical Atlantic (NSA, similar to 16 degrees N-9 degrees N) in April-May 2008. N-2-fixation and primary production were measured as N-15(2) and C-14 uptake, respectively. Dissipation rates of turbulent kinetic energy (epsilon) were measured with a microstructure profiler. The vertical input of nitrate through eddy diffusion was calculated from the product of diffusivity, derived from epsilon, and the gradient of nanomolar nitrate concentration across the base of the euphotic zone. The mean N-2-fixation rate in EA was 56 +/- 49 mu mol N m(-2) d(-1), whereas SSA and NSA had much lower values (similar to 10 mu mol N m(-2) d(-1)). Because of the large spatial variability in nitrate diffusion (34 +/- 50, 405 +/- 888, and 844 +/- 1258 mu mol N m(-2) d(-1) in SSA, EA, and NSA, respectively), the contribution of N-2-fixation to new production in the SSA, EA, and NSA was 44% +/- 30%, 22% +/- 19%, and 2% +/- 2%, respectively. The differences between SSA and NSA in the contribution of N-2 fixation were partly due to the contrasting seasonal forcing in each hemisphere, which likely affected both N-2 fixation rates and vertical nitrate diffusion. The variability in the nitrogen budget of the Atlantic subtropical gyres was unexpectedly high and largely uncoupled from relatively constant phytoplankton standing stocks and primary production rates.